WO2015070227A1

WO2015070227A1 - Location of actor resources

Info

Publication number: WO2015070227A1
Application number: PCT/US2014/065040
Authority: WO
Inventors: Jonathan Paul THOMPSON; Charles Beyer
Original assignee: Amazon Technologies, Inc.
Priority date: 2013-11-11
Filing date: 2014-11-11
Publication date: 2015-05-14

Abstract

In an actor system, messages communicated to and from a first actor in a first location of an actor system can be monitored. A message frequency between the first actor and a second actor in a second location of the actor system can be determined. The message frequency can be compared to a threshold message frequency. A determination cars be made whether to move the first actor from the first location based on comparison of the message frequency to the threshold message frequency.

Description

LOCATION OF ACTOR RESOURCES

CROSS-REFERENCE TO RELATED APPLICATIONS

[000t | This application cla ms the benefit of U.S. Patent Application No. 14/076,855, filed November 1 1 , 2013, the disclosure of which is Incorporated herein by reference in its entirety,

[0802] This application is related to Ihe following applications, each of which is hereby incorporated by reference in its entirety: United States Patent Application No. 14/076,718 filed November 1 L 2013, entitled "VIDEO ENCODING BASED ON AREAS OF INTEREST" (Attorney Docket Number: i 01058.8883); United States Patent Application No. 14/076,821 filed November 1 1 , 2013, entitled "ADAPTIVE SCENE COMPLEXITY BASED ON SERVICE QUALITY" (Attorney Docket Number: 101058,0084); United States Patent Application No. 14/077, 127 Hied November 1 1 , 2013, entitled "SERVICE FOR GENERATING GRAPHICS OBJECT DATA" (Attorney Docket Number: 101058,0086); United States Patent Application No. 14/077, 136 tiled November 1 1 , 2013, entitled "IMAGE COMPOSITION BASED ON REMOTE OBJECT DATA" (Attorney Docket Number: 101058.0087); United States Patent Application No. 14/077, 165 filed November 1 1 , 2013, entitled " ULTIPLE PARALLEL GRAPHICS PROCESSING UNITS" (Attorney Docket Number: 101058.0 ! 10); United States Patent Application No, 14/077,084 filed November ! I , 2013, entitled "ADAPTIVE CONTENT TRANSMISSION" (Attorney Docket Number: 101058,01 1 ); United States Patent Application No. 14/077, 180 Hied November 1 I , 2013, entitled "VIEW GENERATION BASED ON

SHARED STATE" (Attorney Docket Number; 101058.01 15); United States Patent Application No. 14/077, 1 86 filed November 1 1 , 2013, entitled "MULTIPLE STREAM CONTENT

PRESENTATION" (Attorney Docket Number: 101058.01 16); United States Patent Application No. 14/077, 149 filed November 1 1 , 2013, entitled "DATA COLLECTION FOR MULTIPLE VIEW GENERATION" (Attorney Docket Number: 101058.0124); United States Patent Application No. 14/077, 142 filed November 1 L 2013, entitled "STREAMING GAME SERVER VIDEO RECORDER" (Attorney Docket Number: 101058.0125); United States Patent

Application No. 14/077, 146 tiled November 1 1 , 2013, entitled "SESSION IDLE

OPTIMIZATION FOR STREAMING SERVER" (Attorney Docket Nu ber: 101058.0129); United States Patent Application No. 14/077,023 filed November 1 1 , 2013, entitled

"APPLICATION STREAMING SERVICE" (Attorney Docket Number: 101058,0139); United States Patent Application No. 1 902,740 filed November 1 h 2013, entitled "EFFICIENT BANDWIDTH ESTIMATION" (Attorney Docket Number; 101058.0141 ).

BACKGROUND

[0003] A computer computational environment can be set up as an actor system with programmed actors operating concurrently with respect to each other. Messages can be sent between actors to update a state of one of the actors, to request information abou one of the actors, to create new actors and the like. The actors can operate independently of each other and the messages sent between actors can alter the way that the actors operate. The actor system can be hosted by a single computing device or hosted in a distributed system over multiple computing devices,

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Throughout the drawings, reference numbers may he re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate example embodiments described herein and are not intended to limit the scope of the disclosure,

[0005] FIG. 1 depicts an example of an actor system operating in a peer-to-peer computing environment.

[0006] FIGS, 2A and 2B depict examples ofa computing environment with a server- based actor system.

0ΘΘ7 FIG, 3 depicts an embodiment of an actor server system that includes multiple servers in multiple server racks,

[@00§] FIG, 4 depicts an actor server system that Includes servers that can

communicate with each other via communication link(s).

[0009J FIGS. 5A, 5B, SC and 5Ό depict an actor server system and examples of relocating actors within the actor server system.

[0010] FIGS. 6A and 6B depict an example of moving an actor from one location to »other when multiple destination locations are available.

[0011] FIG. 7 depicts a method that can be used by a management service to attempt to move one or more actors to another location.

[0012] FIG. 8 depicts a method of determining whether to move one or more actors after receiving degree of closeness of two actors. |0S13| FIG, 9 depicts a method of determining a degree of closeness for a first actor and a second actor.

[®S1 | FIG. 10 depicts a diagram illustrating an example computing system that may be used in some embodiments.

[0015] FIG. 1 1 depicts a diagram illustrating an example computing system that may be used in some embodiments,

DETAILED DESCRIPTION

[0016] An actor syslem in computer science is a mathematical model of concurrent computation that treats "actors" as the universal primitives of concurrent digital computation: in response to a message that it receives, an actor can make local decisions, create more actors, send more messages, and determine how to respond to the next message received, it has been used both as a framework for a theoretical understanding of computation, and as the theoretical basis for several practical implementations of concurrent systems.

[0017 [ The actor model typically operates under the notion that everything is an actor. This is similar to the philosophy used by some object-oriented programming languages, hut differs in that object-oriented software is typically executed sequentially, while the actors in an actor system can operate concurrently,

]091i] An actor is a computational entity thai, in response to a message it receives, can concurrently, among other actions, send messages to other actors, create new actors, or designate the behavior to be used for the next message it receives. There is no assumed sequence to these actions and they could be carried out in parallel. Messages sent within an actor system may not identify the sender of the message so that the actor system can call for asynchronous communication and control structures as patterns of passing messages. Instead, recipients of messages are identified by address, sometimes called a "mailing address." One actor may be able to communicate only with other actors when it knows the other actors⁵ mailing addresses, An actor can be aware of an address of another actor by obtaining the address from a message, by creating the other actor, and the like.

[001 [ The actor system can include concurrent computation within and among actors, dynamic creation of actors, inclusion of actor addresses in messages, and interaction through direct asynchronous message passing with no restriction on message arrival order. ]ΙΚ)2Θ| An actor system can be used as a framework for modeling, understanding and reasoning about, a wide range of concurrent systems. In one example, an email system can be modeled as an actor system. Accounts can be modeled as actors and email addresses as actor addresses. Other aspects of the email system, such as an account's contact fist, an account's settings and the like, can also be modeled as actors separate from the account actor. In another example, web services can be modeled with simple object access protocol endpoints modeled as actor addresses. In another example, programming objects with locks (such as in Java and€#) can be modeled as a seriaiizer if their implementations permit messages to continually arrive. A serialize-r can be an actor defined by a property that it is continually available to the arrival of new messages (i.e., every message sent to a serializer is guaranteed to arrive). In yet another example, testing and test control notation (TTC ), such as TTCM-2 and TTCN-3, can follow actor systems rather closely. In TTCN, an actor can he a test component: either parallel test component (FTC) or main test component (MTC). Test components can send and receive messages to and from remote partners (peer test components or test system interface).

[ΘΘ2Ι] In a further example, a gaming system can be modeled as an actor system. Each of the characters in a gaming system can he an actor. The gaming system can include an environment in which multiple characters operate independently in a common environment, such as in WORLD OF WARCRAFT and other similar games, In this case, the characters can operate concurrently and independently and the corresponding actors (i.e., the programming actors) can be executing concurrently and independently of each other. The gaming system can also include other actors, such as an actor representing an inventory of items associated with each of the characters, an actor representing characteristics of each of the characters (e.g., skills, abilities, health, etc), and the like, While a number of examples of actor systems are described below in the context of a gaming system, it should be understood that the examples can be similarly applied in any actor system,

[ΘΘ22] Because actors in an actor system communicate messages to each other, the actor system will route messages between the actors. As the number of actors in the actor system increases, so too can the number of messages sent between actors. If every actor n an actor system communicated with every other actor in the actor system, a linear increase in the number of actors may resuit in an exponential increase in the number of messages handled within the actor system. The routing of large numbers of messages and the time that messages take to be sent between actors can cause latency in the actor system. [0023] FIG, 5 depicts an e am le of an actor system operating in a pees -to-peer computing en ironment 100. The computing environment 100 includes a number of computing devices 1 10, 120 and 130, The computing environment 100 can include additional computing devices that are not depicted in FIG. 1 , The computing device 1 10 can include one or more actors i 12a to 1 12n; the computing device 120 can include one or more actors 522a to 522n; and the computing device S 30 can include one or more actors 132a to S 32n. The computing devices 1 10, 120, and 130 can be connected to a network 140 via communication links 5 54, i24_s and 134_s respectively. The network 540 can include one or more of the Internet, a local area network (LAN), a wide area network (WAN), a wireless network (e.g., a WiFi network, a cellular data network, etc.), a the like.

[0024 [ The computing device 1 10 includes a management service i l l . The management service 1 1 1 can manage the actors 1 12a to 1 12n located on the computing device I t O, manage the creation of new actors on the computing device ! 10, manage the deletion of actors from computing device 1 10, route communications to and from actors on the computing device 1 1 , transfer actors from the computing device 1 10 to another computing device or server, receive and host actors transferred from other computing devices or servers to the computing device 1 10 and the like. In the depiction shown in FIG, I , computing devices 520 and 530 do not include management services. The actors 122a to I 22n and 132a to 132n on computing devices 120 and 130 may be able to operate and send messages without the assistance of a management service.

|002Sj In peer-to-peer computing environment 100, when actors in different computing devices pass messages between each other, the messages are routed via the network 140, For example, a message sent from actor 1 52a to actor i 22n will be sent from computing device 1 10 to network 140 via communication link 1 14 and then from the network 140 to the computing device 520 via the communication link 124, In another example, a message sent from actor 132a to actor 1 12n will be sent from computing device 130 to network 140 via communication link 134 and then from the network 140 to the computing device 1 10 via the communication link 1 1 , in yet another example, a message sent from actor i i 2a to actor 5 52n will be passed within the computing device 1 10.

00 6] Using the examples from the preceding paragraph, it is apparent that certain pairs of actors can be "closer" to each other than other pairs of actors. In this context, actors are considered "close" or "far" depending on the speed with which messages can be communicated between them. Among the three examples described in the preceding paragraph, actors 5 52a and 1 52n may be considered the closest of the three examples if communicating messages within computing device ! 10 is faster than communicating messages via the network HO, If both of the communication link 1 14 and the communication link 124 are high-speed communication links and communication link 134 is a low-speed communication link, then actors 1 12a and 122n may be closer to each other than actors D2a and I I 2n because messages communicated between actors 1 12a and 122n may be faster via the high-speed communication links 1 14 arid 124 than messages communicated via high-speed communication links ! 14 and low-speed communication link S 34.

[0027] One ssue with the messages sent between actors in the peer-to-peer computing environment 100 is the latency incurred by communicating messages between different computing devices. While delays due to communicating messages across network(s) 1 0 may be on the order of a few seconds, such delays cars be noticeable to users of the various computing devices. Moreover, while only three computing devices 1 10, 120, and 130 are depicted in FIG. 1 , a similar system could include significantly more computing devices, such as hundreds, thousands or more. As actors on a significant number of computing devices need to

communicate with each other, the latency in sending and receiving messages can increase.

Although an actor system operates on the premise that ail of the actors are operating concurrently and independently of each other and messages are passed asynchronously, delays in interactions between actors can be perceptible to users of the computing device and lower the users' experience.

[0S2S] FIG. 2A depicts a computing environment 200 with a server-based actor system. The computing environment 200 includes a number of computing devices 210, 220 and 230. The computing devices 210, 220 and 230 can be operated by different users to interact with the actor system. Each of the computing devices 210, 220 and 230 can be connected to a network 240 via communication links 214, 224 and 234, respectively. While three computing devices 210, 220 and 230 are depicted in FIG. 2, any number of computing devices could be

communicatively coupled to the network 240.

[ 02 1 Ths computing environment 200 also includes an actor server system 250 in communication with network 240. The actor server system 250 can include one or more servers that host actors in the server-based actor system. In the specific example depicted in FiG. 2, the actor server system 250 hosts actors 212a to 12n, actors 222a to 222n and actors 232a to 232n, In one example, the actors 212a to 212n, 222a to 222n and 232a to 232n may be associated with users of computing devices 210, 220 and 230, respectively, For example, actors 212a to 212n can be associated with an email account of a user of the computing device 210 i an actor-based email system, actors 212a to 212n can be associated with a character of a user of the computing device 210 in an actor-based gaming system, and the like. In another example, the actors 21 2a to 212n, 222a to 222n and 232a to 232n may not be associated with any one user or computing device, but merely actors in an actor-based system.

[GS30J Placing actors of an actor system within an actor server system can reduce latency in sending messages betxveen the actors. For exam le, with actors 2 ί 2a to 212n, 222a to 222n and 232a to 232n hosted by the actor server system 250, messages sent between any two of the actors 212a io 2! 2n, 222a to 222n and 232a to 232rt will be sent within the actor server system 250. While it may take some time to send and receive messages between actors in the actor server system 250, the latency from sending messages can be reduced. For example, messages sent by actors within the actor server system 250 do not need to be sent via the network 240, in this way, the speed of sending messages is not limited by the speed of any

communication link with the network 240 or the speed of communicating within network 240,

|G03i] While all of the messages sent between actors 212a to 212n, 222a to 222n and 232a to 232n can be sent within the actor server system 250, computing devices 210, 220 and 230 can still interact with the actor system. For example, in the case where the actor system is a gaming system and one of the actors hosted by the actor server system 250 may be associated with a character of a user of computing device 210, The user may input controls and commands into the remote computing device which are communicated from the computing device 210 to the actor server system 250 via the network 240. The actor server system 250 can modify the actor associated with the user's character based on those controls and commands. The actor server system 250 can also modify other actors based on those controls and commands (e.g., modify an actor associated with an inventory of the user's character, modify an actor associated with another character that is interacting with the user's character, etc.). Similarly, the actor server system 250 can send back information about the actor system environment back to the computing device 210. The Information sent back to the computing device 210 from the actor server system 250 can include information for the computing device 210 to be able to render updated actors properly (e.g., updates in particular characters, changes in a character's inventory, etc.). The actor server system 250 may also perform all of the processing needed to update the environment from the perspective of the user, and the information sent back to the computing device 210 from the actor server system 250 can include information for the computing device 210 to render the environment to the user. [S032] FIG. 2B depicts anther embodiment of the computing environment 200 with a server-based actor system. The example in FIG. 2B includes actors 212 , 222z and 232?. located on computing device 210, computing device 220 and computing device 230, respectively, and a security service 260 located between the actor server system 250 and the network 240, While the example depicted includes one actor on each of computing devices 210, 220 and 230, the computing devices 210, 220 and 230 could include any number of actors. The actors 212z, 222z and 232 located on computing devices 210, 220 and 230 cars send messages to and receive messages from the actors in the actor server system 250.

[08331 The security service 260 can ensure that messages passed between actors across the network 240 are communicated in a secure fashion, such as by encrypting the messages using a security protocol. The security service 260 can also ensure thai the messages sent by actors on the computing devices 210, 220 and 230 do not perform an actions that are not allowed by the actor system, Because the actors on the computing devices 210, 220 and 230 are not under the control of the actor server system 250, there could be concern that those actors could be programmed to perform malicious functions or other functions that would be improper. For example, In an gaming scenario where characters can earn money, an actor 212z on computing device 210 may send a message to an actor in the actor server system 250 that would attempt to give a character money that was not earned by that character. Since it would be improper to for the actor 212z to perform such a money-transfer function, the security service 260 can intercept such messages to prevent the money-transfer function from occurring, in the depiction of FIG. 2B, the security services is located outside of the actor server system 250, the security service 260 could also he hosted inside of the actor server system. In this case, the security service 260 located inside of the actor server system 250 could monitor and secure messages coming out of and Into the actor server system 250 to and from actors on computing devices 210, 220 and 230,}0034] FIG, 3 depicts an embodiment of an actor server system 300 that includes multiple servers in multiple server racks. The actor server system 300 includes a server rack 3 10 and a server rack 350. Server racks 310 and 350 can be a structure that can accommodate a number of servers in a modular manner. One or more servers can be added to or removed from server racks 310 and 350 without disrupting operation of the other servers in the server racks 10 and 350, A server rack can provide a number of functions to each of the servers in the server rack, such as providing power, cooling, networking communication links and the like. Some advantages to using a server rack include the ability to provision additional server(s) in a server rack if demand for computing resources increases and the ability to remove server(s) from the server rack if demand for computing resources decreases.

[S03S] In the particular embodiment shown in F!G. 3, the server rack 3 10 inciisdes servers 320, 330 and 340, and the server rack 350 inekdes servers 360, 370 and 380. While three servers are depicted in each of server rack 310 and server rack 350, server rack 3 i 0 and server rack 350 could Include any number of servers. Server 320 hosts actors 32 la to 321 rs. server 330 hosts actors 33 J a to 33 I n, server 340 hosts actors 341 a to 34 I n, server 360 hosts actors 361a to 36 ! n, server 370 hosts actors 37i a to 371 n and server 380 hosts actors 381 a to 381 n. The actors 32 \ , 331 , 34 i , 36 I , 371 and 381 may all be part of the same actor system. While the actors 32 S , 331 , 341, 361 , 37 i and 381 could ail be hosted by the same server, It maybe desirable to host the actors 321 , 331 , 341 , 361. 371 and 381 on different servers, such as in the way depicted in FIG, 3, It may be that the number of actors 321, 331, 341 , 361 , 371 and 381 is too great to be hossed by a single server and that multiple servers can be used, such as in the way- depicted in FIG. 3.

[§§36! The actor server system 300 can also include a number of communication links. Server rack 310 includes a number of intra-rack communication 0nk(s) 3 1 1 that permit communication between each of the servers 320. 330 and 340 that are located on the server rack 3 1 , Server rack 350 includes a number of intra-rack communication iink(s) 351 that permit communication between each of the servers 360, 370 and 380 that are located on the server rack 350, The actor server system 300 also inciudes inter-rack communication, iirjk(s) 301 that permit communication between different server racks and permit communication between the server racks and an external network 390, The external network 390 can include a locai area network (LAN), the Internet or any other type of network, in some embodiments, actors in the actor system can also be hosted by servers that are outside of the actor server system 300. In this ease, actors can be hosted by an external system 395. The external; system 393 can be a computing device, server or another actor server system that is In communication with the external network 390.

[00371 Communicating messages between two actors in the actor system shown in FIG. 3 may take different amounts of time depending on the network location between the two actors. For example, a message can be sent between actors in the same server at a first rate. Messages sent within a server, such as a message sent from actor 32 l a to actor 32 I n may be sent at the first rate. In another example, intra-rack communication iink(s) 31 1 may allow communications at a second rate, such as a rate of 1 0 gigabits per second (e.g., via a 100 Gigabit Ethernet ( ! OOGbE) link). The second rate can be slower than the first rate. Messages sent via the intra-raek communication link(s) 31 1 , such as a message sent from actor 321a to actor 331 , can be communicated at the second rate, in another example, inter-rack communication iink(s) 301 may allow communications at a third rate, such as a rate of !O gigabits per second (e.g., via a 10 Gigabit Ethernet (i OGbE) link). The third rate can he slower than the second rate. Messages sent via the inter-rack communication link(s) 301 , such as a message sent from actor 321 to actor 361 a, can be communicated at the third rate. In another example, externai network 390 may allow communications at a fourth rate. The fourth rate can be s ower than the third rate, Messages sent via the external network 390. such as a message sent from actor 321 a to an actor in externai system 395, can be communicated at the fourth particular rate.

[0038] Using the examples from the previous paragraph, the closeness of one actor to other actors can be understood. Closeness of one actor to other actors can be determined based on the time taken to communicate messages from the one actor to the other actors or the rate of communication between one actor and the other actors. The actor 321 a is closest to the other actorfs) on server 320, such as actor 32 i n, because communications between the actor 32 la and the other aetor(s) on server 320 occur at the fastest rate in this example, The actor 32 l a is next closest to the actor(s) on other server(s) within the server rack 310. such as actor 33 la on server 330 and actor 341a on server 340, because communications between the actor 321 and the other actor(s) on other server(s) within the server rack 310 occur at the second Fastest rate in this example. The actor 321a is next closest to the acior(s) on other server(s) within other server rack(s) within the actor server system 300_s such as actor 361 a on server 360 and actor 371 a on server 370 in server rack 350, because communications between the actor 321 a and the other actor(s) on other server{s) within other server raek(s) within the actor server system 300 occur at the third fastest rate in this example. The actor 321 a Is furthest from the actor(s) on the external system 395 because communications between the actor 321 a and actor(s) on the externai system 395 occur at the slowest rate in this example,

1003 1 The closeness of actors to each other and the differences in communication speeds are not limited to those depicted in FIG, 3, Many other configurations could place actors near or close to each other. In one example, actors could be located in different rooms of a data center and actors in different rooms of a data center couid be further apart than actors in different server racks in the same room. In another example, actors could be located in different data centers which couid be further apart than actors in different rooms within a single data center, in yet another example, actors could be located in data centers in different availability zones which eouid be (briber apart than actors in different data centers within a single availability zone. Many other conditions could exist that would place actors near or far from each other.

[ Θ4Θ] FIG. 4 depicts an actor server system 400 that includes servers 410, 420, and 430 that can communicate with each other via communication Iink{s) 440, The communication l nks 440 shown in FIG. 4 show a fairly sim le topology with one connection between each pair of servers 410, 420_s and 430. However, such a topology could be difficult to implement should additional servers be added to the system 400. Many other topologies could be used to implement communication links 440, and communication links 440 are not limited to this one example depicted in FIG. 4. For example, communication links 440 could have a hierarchical topology or a ring topology. The communication links 440 could also be replaced by a etwork_s such as a local area network (LAN), or a communication hub, such as a WiFi router.

[©©41 J The server 410 includes a management service 41 1 and actors 412-414, The management sendee 4 1 I can manage the actors 12-414 located within server 410, manage the creation of new actors within server 410, manage the deletion of actors from server 410, route communications to and from actors on the server 410, transfer actors from the server 410 to another server, receive and host actors transferred from other servers to the server 410 and the like. The server 420 includes a management service 421 and actors 422-424, The server 430 includes a management service 431 and actors 432-434. The management services 423 and 431 can provide similar functionality for their respective servers 420 and 430, as was described with respect to the functionality of management service 41 1 for its server 410.

[0042] The management services 41 1 , 42 f and 431 can maintain a log of the messages sent from and received by the actors on their respective servers 4 S O, 420 and 430. For example, management service 41 1 can maintain a log that includes an indication of the number of t imes that actor 412 has received a message from each of the actors 4 13, 414, 422-424 and 432-434, and the number of times that the actor 412 has sent a message to each of the actors 413, 414, 422-424 and 432-434. Such a log can be used to determine a message frequency of messages sent between the actor 412 and each of the other actors in the actor server system 400. A message frequency can be based on the overall number of messages sent during a particular period of time, based on sizes of messages sent during a particular period of time, or based on a combination of the number of overaii messages and the sizes of messages sent during a particular period of time. Message frequencies can indicate how "talkative" the actor 412 is with the other actors in the actor server system 400. If two actors are more talkative with each other than with other actors, it may be advantageous to place the two actors closer to each other within the actor server system 400 to reduce the time that messages take to pass between the two tentative actors.

JMM3J Based on the message frequencies of messages being sent between actors, a degree of closeness can be determined tor two or more actors. A degree of closeness can be a binary option (e.g., either actors should be close or actors do not need to be close), a range of values (e.g., a range of value from 1 to SO where 1 is associated with a lowest need for the actors to be close and 10 is associated with the highest indicator that the actors shouid be close), and/or any other indication of a degree, in another embodiment, a degree of closeness for two or more actors can he set by a developer of the actor system hosted by actor server sysEem 400, by an operator of the actor system hosted by actor server system 400, or by any other group or individual As is discussed in greater detail below, the degree of closeness—whether determined based on a message frequency of messages sent or set by a person or grou — can be used to locate one or more actors within the actor server system 4ϋΰ,

[0044] FIGS. 5 A to 5D depict an actor server system 500 and examples of relocating actors within the actor server system 500, The actor server system 500 includes servers 5 10, 520 and 530 that are in communication with each other via communication lmk{s) 540, Servers 510, 520 and 530 include management services 51 1 , 512 and 513, respectively. At the time depicted in FIG, 5A, server 510 hosts actors 512-519, server 520 hosts actors 522 and 523 and server 530 hosts actors 532-535, The management service 51 i may determine to move actors 518 and 5 f 9 to one or more of the other servers 520 and 530. Such a determination may be made on one or more of a number of factors,, such as server 510 reaching or nearing a capacity of actors, management service 51 1 determining that aetors 518 and 51 are more talkati e with actors that are not on server 5 10 than with actors on server 51 or the like. Each of the times depicted in FIGS. SB to 5D represents a different way that management server 51 1 can move actors 518 and 519 from the server 510 to one or more of the other servers 520 and 530,

[QMS] FIG. SB depicts a load-balancing approach to move actors 5 i 8 and 519 from the server 510. At the time depicted In FIG. 5 A, management service 5 i 1 may determine to attempt to move actors 5 18 and 51 from server 510 to another server. This determination may be made because nsage of computing resources in server 510 are at or nearing a maximum level, because the message frequency between each of actors 5 18 and 51 and other actors outside of server 510 are above a threshold, or for any other reason.

f§04 ] Management server 5 1 1 can send an availability inquiry to each of management service 521 and management service 531 requesting an availability of the server 520 and the server 530, respectively, io hast additional actors. Responses from the management service 521 and management service 531 can be sent to management service 5 i i indicating a level of availability of the server 520 and the server 530 to host additional actors, in the particular embodiment shown in FIG. 5 A, based on the responses from management service 521 and from management service 53 1 to management service 51 1 can indicate that both the server 520 and the server 530 have availability to host additional actors. The responses can also indicate that tbe server 520 has greater availability than server 530 io host additional resources, Sn the particular instance shown in FIG, 5B, the management service 51 5 sent the actors 5 18 and 51 to the server 520. Such a decision may be made based_f at least In part, on an intent to balance the load among the servers 510, 520 and 530.

[0047] FIG. SC depicts an approach to move actors 518 and 519 from the server 5 10 based on message frequencies between actors. At the time depicted in FIG. 5A, management service 51 1 may determine that actors 5 18 and 5 19 are sending messages to actors outside of server 5 10 more frequently that they send messages to actors within the server 510. For example, management service 5 1 1 can determine that actors 51 S and 519 send messages to actors located on server 530 at a message frequency that is above a threshoid message frequency. Because the message frequency between actors 518 and 519 and actors located on server 530 Is above the message threshold frequency, the management service 51 1 can send an availability inquiry to management service 5 1 and the management service 531 can respond with an indication thai server 530 has availability to host additional actors. In the particular instance shown in FIG. 5C, the management service 51 1 sent the actors 518 and 19 to the server 530,

[@048] FIG. 5D depicts an approach to move actors 518 and 51 from the server 5 i 0 based on a degree of closeness received by the actor server system 500. The actor server system 500 can receive an indication of a degree of closeness for actor 5 I S and a degree of closeness for actor 5 19, The degree of closeness for actor 518 can indicate that actor 5 I S should be located close to actor 522. The degree of closeness for actor 519 ean indicate that actor 519 should be located close to actor 532, The degrees of closeness for actors 18 and 519 can be submitted by a developer of the actor system, an operator of the actor system, or any other group or individual. Based on the degree of closeness for actor 518, management service 51 1 can send an availability inquiry to management service 521 about an availability of the server 520 to host actor 1 8 and the management service 521 can respond with an indication that server 520 has availability to host actor 518, Based on the degree of closeness for actor 519, management service 5 1 1 can send an availability inquiry to management service 531 about an availability of the server 530 to host actor 5 I 9 and the management service 531 can respond with an indication that server 530 has availability to host actor 51 . irs the particular instance shown in FIG. 5D, the management service 1 1 sent the actor 518 to the server 520 and the management service 51 1 sent the actor 519 to the server 530.

[0049] FIGS. 6A and 6B depict an example of moving an actor from one location to another when multiple destination locations are available. FIGS. 6A and 6B depict an actor server system 600 that includes a server rack 610 and a server rack 640. The server rack 610 includes server 620 and server 630. Server 620 includes a management service 621 and hosts actors 622-627, Server 630 includes a management service 631 and hosts actors 632 and 633. The servers 620 and 630 within server rack 10 can communicate with each other via intra-raek communication link(s) 61 \ . The server rack 640 includes server 650 and server 660, Server 650 includes a management service 651 and hosts actors 652-657. Server 660 includes a

management service 661 and hosts actors 662-664, The servers 650 and 660 within server rack 640 cart communicate with each other via intra-rack communication iink(s) 641 .

Communications between server racks 610 and 640 can be carried via inter-rack communication iink(s) 601. The intra-rack communication link(s) 61 1 and 641 may communicate messages at a faster rate than the inter-rack communication link(s) 601 may communicate messages.

[θ®50| At the time depicted in F!G. 6A, the management service 621 may determine that actor 627 should be moved to another server. The determination that actor 627 should be moved can be based on one or more of server 620 nearing a capacity of hosted actors, a degree of closeness of actor 627 to another actor being determined, a degree of closeness of actor 62? to another actor being received and the like. The management service 621 can send an availability inquiry to each of the other management services 631, 651 and 661 in the actor server system 600. At the particular time depicted in FIG. 6A, the management service 651 may respond with an indication that the server 650 is unavailable to host another actor and the management services 63 1 and 661 may respond with an indication that the servers 630 and 660 are available to host another actor.

10051} The determination that actor 627 should be moved can be based can also be based on a determination of optimization of computing resources within the actor system, in one embodiment, if computing resources within the actor system are underutilized, it may be more optimal for the underutilized computing resources to be shut down instead of moving actor 627. For example, it may be possible to move the actors 632 and 633 on server 630 to server 660 and remove server 630 from service, in this embodiment, the termination may be made to ieave actor 627 on server 620 so that server 630 could be removed to save computing resources. In another embodiment, the determination of opt mization of computing resources within the actor system cars be based on a local, regional or global view of the actor system. A global view of the actor system cars provsde a complete analysis of the actor system, but it may take a significant about of time to analyze the entire actor system. A local view of the actor system ean provide an analysis of the actors close to one or more actors; however, a focal analysis may not provsde enough information about the actor system. A regional view may provide an analysis that is somewhere between the local view and the giobal view of the actor system. One way to approximate a global view of the actor system would be to combine multiple regional views of the actor system to approximate the giobai view of the actor system. Such an approximation of the giobal view of the actor system may not take as much time to perform as performing a full giobai view of the actor system.

[0052] The management service 621 may determine which of ibe available servers 630 and 660 to which the actor 627 should be transferred based on the degree of closeness between actor 62? and another actor in the actor server system 600. For example, the management service 621 may be aware of a degree of closeness between actor 627 and actor 656 on server 650. The degree of closeness between actor 627 and actor 656 can be based on a message frequency of messages sent between actor 627 and actor 656, based on a received degree of closeness between actor 627 and actor 656, or based on any other information, Placing the actor 627 on server 650 would place the actor 627 in the closest location to actor 656. However, server 650 is not available to host actor 627, Between the available servers 630 and 660, the management service 621 can determine that the actor 627 will be closer to actor 656 if it is moved to server 660 than if it is moved to server 630 because the intra-raek communication Hnk(s) 641 communicate messages between server 660 and server 650 at a faster rate than the inter-rack communication iirsk(s) 601 communicate messages between server 630 and server 650, At the time shown in FIG. 6B, the actor 627 has been moved to server 660.

|0853] FIG. 7 depicts a method 7QQ that can be used by a management service to attempt to move one or more actors to another location. At block 701 , the management service can determine that one or more actors should be moved to another location. As discussed above, such a decision can be based on a usage of a computing device or server on which the one or more actors are hosted, based on a degree of closeness of the one or more actors, or based on any other information or reason. The decision can also be based on a determination of optimization of computing resources within the actor system. At block 702, an inquiry can be made about

~ \ 5 - availability of other locations to accept the one or more actors. The inquiry can include an inquiry message sent from one management service on a server or computing device to another management service on another server or computing device, At block a 703, a determination can be made as to how many locations are available to host the one or more actors.

[0Θ54] If, at block 703, it is determined that there are no locations available to host the one or more actors, then, at block 704, the one or more actors can be left in their original location. At block 705, a signal can be sent that additional locations may need to be provisioned. Such a signal can include one or more of a signal sent to a developer or operator of the actor system to purchase additional computing resources available in a server system, a signal for a network technician to install an additional server within a server rack, a signal to automatically provision an additional server to host actors and the like. If additional locations are provisioned, the one or more actors can then be transferred to the newly-provisioned locations,

[0055] if, at block 703, it is determined that only one location is available, then, at block 706, the one or more actors can be sent to that available location, In an embodiment not depicted in FIG. 7, an additional decision can be made after block 703 and before block 703, In that embodiment, a decision can be made whether the one or more actors will be closer to other actors with which the one or more actors frequently send or receive messages, if the one or more actors would be closer to those other actors at the available location, then the method could proceed to block 706 where the one or more actors are sent to the available location. However, if the one or more actors would not be closer to those other actors at the available location, then the one or more actors can be left at their original location.

|805β| If, at block 703, it is determined that more than one location is available, then, at block 707, a degree of closeness can be determined between the one or more actors and other actors in the actor system. At block 708, a destination location from the available locations can be determined based at least In part on the degree of closeness. The destination location may not be the closest location of the one or more actors to other actors with which the one or more actors frequently send messages, but the destination location may be the closest available location. Additionally, in the case where there are more than one actors being moved, more than one destination location may be determined, At block 709, the one or more actors are sent to the destination location determined at block 708.

|@057] FIG, S depicts a method 800 of determining whether to move one or more actors after receiving degree of closeness of two actors. At bioek SO 1 , an indication of a degree of closeness between two actors can be received. The degree of closeness can be a binary option (e.g., either actors should be close or actors do not need to be close), a range of values (e.g., a range of value from I to 10 where I is associated with a lowest need for the actors to be dose and 10 is associated with the highest indicator that the actors should be close), and/or any other indication of a degree, For example, the indication of the degree of closeness can indicate that a first actor and a second actor should be located close to each other within an actor system. At block 802, an inquiry can he made whether any possible new locations for the first and/or second actor exist. A new location can be a server, a server rack, a computing device and the like.

[©058] At block 803, a determination can be made whether any additional locations are available to host one or both of the first and second actors. The determination can he made based on responses to the inquiries sent at block 802, If, at block §03, it is determined that, no new locations are available to host one or both of the first and second actors, then, at block 804, a signal can be raised regarding a possible need for additional locations. Additional locations can be made by provisioning an additional server, by purchasing access to additional resources within an actor server system and the like. At block 805, the first and second actors can be left in their original locations,

[0859] If, at block 803, it is determined that one or more new locations are available to host one or both of the first and second actors, then, at block 806, a determination is made whether the possible one or more new locations would permit the first and second actors to be located closer to each other. For example, if one of the possible new locations is a server that already hosts the second actor, then placing the first actor on the same server would locate the first and second actors closer to each other. In another example, if the first and second actors are located on different servers within the same server rack and the only possible new location is a server in another server rack, then moving one of the first and second actors to the server in the other server rack would not locate the first and second actors closer to each other, If. at block 806, it is determined that the possible new iocations would not locate the first and second actors closer to each other, then the method can proceed to block 80S where the first and second actors are left in their original locations. However, if, at b ock 806, it is determined that the possible new locations would locate the first and second actors closer to each other, then, at block 807. one or both of the first and second actors can be moved to one or more of the new iocations.

[006lf| FIG. 9 depicts a method 900 of determining a degree of closeness for a first actor and a second actor. At block 901 , the addresses of messages sent from and received by a first actor ca be monitored. Monitoring the messages can include creating a log of the destinations of each message sent by the first actor and the origins of each message received by the first actor, At block 902, a message frequency sent between the first actor and a second actor can be determined, The message frequency can be based ors the overall number of messages sent between the first actor and a second actor during a particular period of time, based on sizes of messages sent between the first actor and a second actor during a particular period of time, or based on a combination of the number of overall messages and the sizes of messages sent between the first actor and a second actor during a particular period of time. The message frequency cars be determined based on information from a log created during the monitoring in block 902, At block 903. a determination can be made whether the message frequency exceeds a threshold. The threshold can be a static threshold, such as a predetermined frequency. The threshold can also be a variable threshold, such as a threshold based on a message frequency between the first actor and other actors in the same server.

S¾061] If, at block 903, il is determined that the message frequency between the first actor and the second actor exceeds the threshold, then, at block 904, a degree of closeness between the first and second actors can be determined. The degree of closeness between the first and second actors can be based on the message frequency between the first and second actors. At block 905₍ a determination can be made, based at least in part on the degree of closeness between the first and second actors, whether one or both of the first and second actors should be moved to a new location. However, if, at block 903, it is determined that the message frequency between the first actor and the second actor docs not exceed the threshold, then, at block 906, the first and second actor can be left in their original locations,

[0 62f FIG. 10 illustrates an example computing environment in which the

embodiments described herein may be implemented, FIG. 10 is a diagram schematically illustrating an example of a data center 1010 that can provide computing resources to users 2000a and 1 00b (which may be referred herein singularly as user 1000 or in the plural as users 1000) via user computers 1002a and 1 02b (which may be referred herein singularly as computer 1002 or In the plural as computers 1002) via a communications network 1030, Data center 1010 may be configured to provide computing resources for executing applications on a permanent or an as-needed basis, The computing resources provided by data center 1010 may include various types of resources, such as gateway resources, load balancing resources, routing resources, networking resources, computing resources, volatile and non-volatile memory resources, content delivery resources, data processing resources, data storage resources, data communication resources and the like. Each type of computing resource may be general-purpose or may be available in a number of specific configurations. For example, data processing resources may be

- Ϊ 8 - available as virtual machine instances ibat may be configured io provide various web services. In addition, combinations of resources may be made available via a network and may be configured as one or more web services. The instances may be configured to execute applications, including web services, such as application services, media services, database services, processing services, gateway services, storage services, routing services, security services, encryption services, load balancing services, application services and the like. These web services may be configurable with set or custom applications and may be configurable in size, execution, cost, latency, type, duration, accessibility and in any other dimension. These web services may be configured as available infrastructure for one or more clients and can include one or more applications configured as a platform or as software for one or more clients, These web services may be made available via one or more communications protocols. These communications protocols may include, for examp e, hypertext transfer protocol (HTTP) or non- HTTP protocols. These communications protocols may also include, for example, more reliable transport layer protocols such as transmission control protocol (TCP) and less reliable transport layer protocols such as user datagram protocol (UDF). Data storage resources may include file storage devices, block storage devices and the like,

[0063] Each type or configuration of computing resource may be available in different sizes, such as large resources— consisting of many processors, large amounts of memory and/or large storage capacity—and small resources—consisting of fewer processors, smaller amounts of memory and/or smaller storage capacity. Customers may choose to allocate a number of small processing resources as web servers and/or one large processing resource as a database server, for example.

§0ΰό4] Data center 1010 may include servers 1016a-b (which may be referred herein singularly as server 1016 or in the plural as servers 1016) that provide computing resources. These resources may be available as bare metal resources, or as virtual machine instances 1018a- d and (which may be referred herein singularly as virtual machine instance 1018 or in the plural as virtual machine instances 101 8), Virtual machine instances 1018c and 1018d are interest virtual machine instances. The interest virtual machine instances 1018c and 10 Ϊ 8d may be configured to perform ail or any portion of the encoding techniques based on areas of interest in accordance with the present disclosure and described in detail below. As should be appreciated, while the particular example illustrated in FIG. 10 includes one interest virtual machine in each server, this is merely an example, A server may include more than one interest virtual machine or may not include any interest virtual machines. [0065] The availability of visualization technologies for computing hardware has provided benefits for providing large scale computing resources for customers and allowing computing resources to be efficiently and securely shared between multiple customers. For example, visualization technologies may aiiow a physical computing device to be shared among multiple users by providing each user with one or more virtual machine instances hosted by the physical computing device, A virtual machine instance ma be a software emulation of a particular physical computing system that acts as a distinct logical computing system. Such a virtual machine instance provides isolation among multiple operating systems sharing a given physical computing resource. Furthermore, some visualization technologies may provide virtual resources that span one or more physical resources, such as a single virtual machine Instance with multiple virtual processors that spans multiple distinct physical computing systems.

]0066J Referring to FIG. 10, communications network 1 30 may, for example, be a publicly accessible network of linked networks and possibly operated by various distinct parties, such as the Internet, In other embodiments, communications network 5030 may be a private network, such as, a corporate or university network that is wholly or partially inaccessible to non-privileged users. In still other embodiments, communications network 1030 may include one or more private networks with access to and/or from the Internet,

[0067] Communication network 1 30 may provide access to computers 1002, User computers 1002 may be computers utilized by users 1000 or other customers of data center 1010, For instance, user computer 1002a or 1002b may be a server, a desktop or laptop personal computer, a tablet computer, a wireless telephone, a personal digital assistant (PDA), an e-book reader, a game console, a set-top box or any other computing device capable of accessing data center 1010, User computer 5002a or 1 02b may connect directly to the Internet (e.g., via a cable modem or a Digital Subscriber Line (DSL)). Although only two user computers 1002a and 1002b are depicted, it should be appreciated that there may be multiple user computers.

[OOfiSj User computers 1002 may also be utilized to configure aspects of the computing resources provided by data center 1010, In this regard, data center 1010 might provide a gateway or web interlace through which aspects of its operation may be configured through the use of a web browser application program executing on user computer 1002, Alternately, a stand-alone application program executing on user computer 1002 might access an application programming interface (API) exposed by data center 1010 for performing the configuration operations. Other mechanisms for configuring the operation of various web services available at data center 1010 might also be utilized. [0069] Servers 1016 shown its FIG, S O may be standard servers configured

appropriately for providing the computing resources described above and may provide computing resources tor executing one or more web services and/or applications, in one embodiment, the computing resources may be virtual machine instances 1038. Ire the example of virtual machine instances, each of the servers 1016 may be configured to execute an instance manager 1020a or 1020b (which may be referred herein singularly as instance manager 1020 or in the plural as instance managers 1020) capable of executing the virtual machine instances 1018. The instance managers 1020 may be a virtual machine monitor (VMM) or another type of program configured to enable the execution of virtual machine instances 1018 on server 1016. for example. As discussed above, each of the virtual machine instances 1018 may be configured to execute ail or a portion of an application.

J 070] It should be appreciated that although the embodiments disclosed above discuss the context of virtual machine instances, other types of implementations can be utilized with the concepts and technologies disclosed herein. For example, the embodiments disclosed herein might also be utilized with computing systems that do not utilize virtual machine instances.

[0071] In the example data center 1010 shown in FIG. 10, a router 1014 may be utilized to interconnect the servers i 0 i6a and 1016b, Router 1034 may also be connected to gateway 1040. which Is connected to communications network 1030, Router 1014 may be connected to one or more load balancers, and alone or in combination may manage

communications within networks in data center 1010, for example by forwarding packets o other data communications as appropriate based on characteristics of such communications (e.g., header information including source and/or destination addresses, protocol identifiers, size, processing requirements, etc) and/or the characteristics of the private network (e.g., routes based on network topology, etc.), it will be appreciated that, for the sake of simplicity, various aspects of the computing systems and other devices of this example are illustrated without showing certain conventional details, Additional computing systems and other devices may be interconnected in other embodiments and may be interconnected in different ways.

| 0?2] In the example data center 1010 shown in FIG. 10, a server manager 1015 is also employed to at least in part direct various communications to, from and/or between servers 1016a and 101 b, While FIG. 10 depicts router 3014 positioned between gateway 1040 and server manager 1015, this merely a exemplary configuration. In some cases, for example, server manager 1015 may be positioned between gateway 1040 and router 1014, Server manager 1015 may, in some cases, examine portions of incoming communications from user computers 1002 to determine one or more appropriate servers 1016 to receive and/or process the Incoming communications, Server manager ( 0 ( 5 may determine appropriate servers to receive and/or process the incoming communications based on factors such as an identity, location or other attributes associated with user computers 1002, a nature of a task with which the communications are associated, a priority of a task with which the communications are associated, a duration of a task with which the communications are associated, a size and/or estimated resource usage of a task with which the communications are associated and many other factors. Server manager 1055 may, for example, collect or otherwise have access to state information and other information associated with various tasks in order to, for example, assist in managing communications and other operations associated with such tasks.

[0873] It should be appreciated that the network topology illustrated in FIG. 10 has been greatly simplified and that many more networks and networking devices may be utilized to Interconnect the various computing systems disclosed herein. These network topologies and devices should be apparent to those skilled in the art,

[0Θ74] It should also be appreciated that data center 101 described in FIG, 10 is merely illustrative and that other implementation might be utilised. Additionally, it should be appreciated that the functionality disclosed herein might be implemented in software, hardware or a combination of software and hardware. Other implementations should be apparent to those skilled in the art. It should also be appreciated that a server, gateway or other computing device may comprise any combination of hardware or software that can interact and perform the described types of functional ty, including without limitation desktop or other computers, database servers, network storage devices and other network devices, PDAs, tablets, cellphones, wireless phones, pagers, electronic organizers, Internet appliances, television-based systems (e.g,, using set top boxes and/or personal/digital video recorders) and various other consumer products that include appropriate communication capabilities, In addition, the functionality provided by the illustrated modules may in some embodiments be combined in fewer modules or distributed in additional modules. Similarly, in some embodiments the functionality of some of the illustrated modules may not be provided and/or other additional functionality may be available.

[0875] In at least some embodiments, a server that implements a portion or ail of one or more of the technologies described herein may Include a general-purpose computer system that includes or is configured to access one or more computer-accessible media, FIG. 1 ? depicts a generai-parpose computer system that includes or is configured to access one or more computer- accessible media. Irs the illustrated embodiment, computing device 1 100 includes one or more processors 1 1 10a. 1 1 10b and/or S 1 10n (which may be referred herein singularly as "a processor 1 1 10" or in the plural as "the processors 1 1 10") coupled to a system memory 1 120 via an input/output (I/O) interface 1 130. Computing device 1 100 rther includes a network interface 1 140 coupled to I/O interface 1 130.

In various embodiment, computing device 1 100 may be a uniprocessor system including one processor 1 1 10 or a multiprocessor system including several processors 1 1 10 (e.g., two, four, eight or another suitable number). Processors 1 U 0 may be any suitable processors capable of executing instructions. For example, in various embodiments, processors 1 1 30 may be general-purpose or embedded processors implementing any of a variety of instruction se architectures (IS As), such as the x86, PowerPC, SPARC or MIPS ISAs or any other suitable ISA. In multiprocessor systems, each of processors I Π 0 may commonly; but not necessarily, implement the same ISA.

] 00T7J System memory 1 1 0 may be configured to store instructions and data accessible by processor(s) 1 1 10, In various embodiments, system memory 1 120 may be implemented using any suitable memory technology, such as static random access memory

(SRAM), synchronous dynamic RAM (SDRAM), nonvolaiiie/Fiash^¾ ype memory or any other type of memory, In the illustrated embodiment, program instructions and data implementing one or more desired functions, such as those methods, techniques and data described above, are shown stored within system memory 1 120 as code 1 125 and data 1 126.

JW7S] 1ft one embodiment I/O interface 1 130 may be configured to coordinate I/O traffic between processor 1 1 10, system memory 1 120 and any peripherals in the device, including network interface 1 140 or other peripheral interfaces. In some embodiments. I/O interface I I 30 may perform any necessary protocol, timing or other data transformations to convert data signals from one component (e.g., system memory 1 120} into a format suitable for use by another component (e.g., processor 1 1 10). In some embodiments, I/O interface 1 1 30 may include support for devices attached through various types of peripheral buses, such as a variant of the Peripheral Component Interconnect (PCI) bus standard or the Uni versal Serial Bus (USB) standard, for example. In some embodiments, the function of I/O interface 1 130 may be split into two or more separate components, such as a north bridge and a south bridge, for example. Also, in some embodiments some or all of the functionality of I/O interface 1 130, such as an interface to system memory 1 S20_? may be Incorporated directly into processor 1 1 1 0. [S 791 Network interface ί 140 may be configured to allow data to be exchanged between computing device I I 00 and other device or devices 1 160 attached to a network or networks 1 150, such as other computer systems or devices, for example. In various

embodiments, network interface 1 140 may support communication via any suitable wired or wireless general data networks, such as types of Ethernet networks, for example. Additionally, network interface 1 140 may support communication via telecommunications/telephony networks, such as anaiog voice networks or digital fiber communications networks, via storage area networks, such as Fibre Channel SANs (storage area networks), or via any other suitable type of network and/or protocol,

[8Θ8Θ] I n some embodiments, system memory 1 120 may be one embodiment, of a computer-accessible medium configured to store program instructions and data as described above for implementing embodiments of the corresponding methods and apparatus. However, in other embodiments, program, instructions arsd/or data may be received, sent or stored upon different types of computer-accessible media. Generally speaking, a computer-accessible medium may include non-transitory storage media or memory media such as magnetic or optical media, e.g., disk or DVD/CD coupled to computing device 1 100 via I/O interface 1 130. A non- transitory computer-accessible storage medium may also include any volatile or non-volatile media such as RAM (e.g. SDRAM, DDR SDRAM, RDRAM, SRAM, etc.), ROM etc., that may be included in some embodiments of computing device 1 100 as system memory 1 320 or another type of memory. Further, a computer-accessible medium may Include transmission media or signals such as electrical, electromagnetic or digital signals conveyed via a communication medium such as a network and/or a wireless link, such as those that may be implemented via network interface 1 140, Portions or all of multiple computing devices such as those Illustrated in FIG. i 1 may be used to Implement the described functionality in various embodiments for example, software components running on a variety of different devices and servers may collaborate to provide the functionality. In some embodiments, portions of the described functionality may be implemented using storage devices, network devices or special-purpose computer systems, in addition to or instead of being implemented using general-purpose computer systems. The term "computing device," as used herein, refers to at least all these types of devices and is not limited to these types of devices.

[GS81] A compute node, which may be referred to also as a computing node, may be implemented on a wide variety of computing environments, such as commodity-hardware computers, virtual machines, web services, computing clusters and computing appliances, Any of these computing devices or environments may, for convenience, be described as compute nodes.

[00S2] A network set up by an entity such as a company or a public sector organization to provide one or more web services (such as various types of cloud-based computing or storage) accessible via the Internet and/or other networks to a distributed set of clients may be termed a provider network. Such a provider network may nc ude numerous data centers hosting various resource pools, such as collections of physical and/or vitalized computer servers, storage devices, networking equipment, and the like, needed to implement and distribute the

infrastructure and web services offered by the provider network. The resources may in some embodiments be offered to clients in various units related to the web service, such as an amount of storage for storage, processing capability for processing, as instances, as sets of related services and the like, A virtual computing instance may, for example, comprise one or more servers with a specified computational capacity (which may be specified by indicating the type and number of CPUs, the main memory size and so on) and a specified software stack (e.g., a particular version of an operating system, which may in turn ran on top of a hypervisor).

]0083] A number of different types of computing devices may be used singly or in combination to implement the resources of the provider network in different embodiments, including general purpose or special purpose computer servers, storage devices, network devices and the like, in some embodiments a client or user may be provided direct access to a resource instance, e.g., by giving a user an administrator login and password. In other embodiments the provider network operator may allow clients to specify execution requirements for specified client applications and schedule execution of the applications on behalf of the client on execution platforms (such as application server instances, Java™ virtual machines (JVMs), general purpose or special-purpose operating systems, platforms that support various interpreted or compiled programming languages such as Ruby, Perl, Python, C, C-H- and the like or high-performance computing platforms) suitable For the applications, without, For example, requiring the client to access an instance or an execution platform directly. A given execution platform may utilize one or more resource instances in some implementations; in other implementations multiple execution platforms may be mapped to a single resource instance,

[0§84| in many environments, operators of provider networks that implement different types of virtualized computing, storage and/or other network-accessible functionality may allow customers to reserve or purchase access to resources in various resource acquisition modes, The computing resource provider may provide facilities for customers to select and launch the desired computing resources, deploy application components io the computing resources and maintain an application executing in the environment In addition, the computing resource provider may provide further facilities for the customer to quickly and easily scaie up or scale down the numbers and types of resources allocated to the application, either manually or through automatic sealing, as demand for or capacity requirements of the application change. The computing resources provided by the computing resource provider may be made available in discrete units, which may be referred to as instances. An instance may represent a physical server hardware platform, a virtual machine instance executing on a server or some combination of the two. Various types and configurations of instances may be made available, including different sizes of resources executing different operating systems (OS) and/or hypervisors, and with various installed software applications, runtimes and the l ke, instances may further be available in specific availability zones, representing a logical region, a fault tolerant region, a data center or other geographic location of the underlying computing hardware, for example. Instances may be copied within an availability zone or across availability zones to improve the redundancy of the instance, and instances may be migrated within a particular availability zone or across availability zones. As one example, the latency for client communications with a particular server in an availability z e may be less than the latency for client communications with a different server. As such, an instance may be migrated From the higher latency server to the lower latency server to improve the overall client experience.

]80S5] Sn some embodiments the provider network may be organized into a plurality of geographical regions, and each region may include one or more availability zones. An availability zone (which may also be referred to as an availability container) in turn may comprise one or more distinct locations or data centers, configured in such a way that the resources in a given availability zone may be isolated or insulated from failures in other availability zones. That is, a failure in one availability zone may not be expected to result in a failure in any other availability zone. Thus, the availability profile of a resource instance is intended to be independent of the availability profile of a resource instance in a different availability zone. Clients may be able to protect their applications from failures at a single location by launching multiple application instances in respective availability zones. At the same time, in some implementations inexpensive and low latency network connectivity may be provided between resource instances that reside within the same geographical region (and network transmissions between resources of the same availability zone may be even taster). [01)86] Thus, as set forth above, a content provider may provide content to a destination over a network such as the interne! usin , for example, streaming content delivery techniques, A content provider may, for example, provide a content delivery service that may reside on one or more servers, The service may be scalable to meet the demands of one or more customers and may increase or decrease in capability based on the number and type of incoming client requests. The content delivery service may, in some eases, process a content item in parallel across multiple nodes of the content delivery service. T is may be done, in one em od ment, to reduce the latency for rendering the content item. Portions of the content delivery service may also be migrated to be placed in a position of reduced latency with a requesting client. In some cases, the content provider may determine an '"edge" of a system or network associated with the content provider that is physically and/or logically closest to a requesting client. The content provider may then, for example, "spin-up," migrate resources, or otherwise employ components associated with the determined edge for interacting with requests from the client. Such an edge determination process may, in some cases, provide an efficient technique for identifying and employing components that are well suited to interact with a particular client, and may, in some embodiments, reduce the latency for communications between a content provider and one or more clients.

Each of the processes, methods and algorithms described in the preceding sections may be embodied in, and fully or partially automated by, code modules executed by one or more computers or computer processors. The code modules may be stored on any type of non- transitory computer-readable medium or computer storage device, such as hard drives, solid state memory, optical disc and/or the like. The processes and algorithms may be implemented partially or wholly in application-specific circuitry. The results of the disclosed processes and process steps may be stored, persistently or otherwise, in any type of non-transitory computer storage such as, e,g.₃ volatile or non-volatile storage.

[mm] The various features and processes described above may be used independently of one another or may be combined in various ways. Ail possible combinations and

subcombinations are intended to fall within the scope of this disclosure. In addition, certain method or process blocks may be omitted in some implementations. The methods and processes described herein are also not limited to any particular sequence, and the blocks or states relating thereto can be performed in other sequences that are appropriate. For example, described blocks or states may be performed in an order other than that specifically disclosed, or multiple blocks or states may be combined in a single block or state. The example blocks or states may be performed n serial in parallel, or irs some other manner. Blocks or states may be added to or removed from the disclosed example embodiments. The example systems and components described herein may be configured differently than described. For example, elements may be added to, removed from or rearranged compared to the disclosed example embodiments.

[QQ 9] It will a!so be appreciated that various items are illustrated as being stored in memory or on storage while being used, and that these items or portions of thereof may be transferred between memory and other storage devices for purposes of memory management and data integrity. Alternatively, irs other embodiments some or all of the software modules and/or systems may execute in memory on another device and communicate with the illustrated computing systems via inter-computer communication. Furthermore, in some embodiments, some or all of the systems and/or modules may be implemented or provided in other ways, such as at least partially in firmware and/or hardware, including, but not limited to, one or more application-specific integrated circuits (ASICs), standard integrated circuits, controllers (e.g., by executing appropriate instructions, and including microcontrollers and/or embedded controllers), field-programmable gate arrays (FFGAs), complex programmable logic devices (CPLDs), etc. Some or all of the modules, systems and data structures may also be stored (e.g., as software instructions or structured data) on a computer-readable medium, such as a hard disk, a memory, a network, or a portable media article to he read by an appropriate drive or via an appropriate connection. The systems, modules and data structures may also be transmitted as generated data signals {e.g., as part of a carrier wave or other analog or digital propagated signal) on a variety of computer-readable transmission media, including wireless-based and wired/cahie-hased media, and may take a variety of forms (e.g., as part of a single or multiplexed analog signal, or as multiple discrete digital packets or frames). Such computer program products may also take other forms in other embodiments. Accordingly, the present invention may be practiced with other computer system configurations.

[0090] The foregoing may be better understood iu view of the following clauses:

I , An actor system comprising:

a plurality of servers in communication with each other;

a plurality of actors hosted by the plurality of servers; and

a management service located on at least a first server of the plurality of servers, the management service configured to; receive an indication of a degree of closeness between a first actor on the first server and a second actor on a second server of the plurality of servers,

send an availability inquiry to one or more of the plurality of servers, receive a response from the one or more of die plurality of servers, determine a destination server from the one or more of the plurality of servers based on a location of the destination server with respect to the second actor, wherein the determination of the destination server is based on the degree of closeness and a location of t he destination server, and

move the first actor from the first server to the destination server,

2, The actor system of clause 1 , wherein the destination server is one of the second server or another server located in a server rack in which the second server is located,

3, The actor system of clause 1 , wherein the management service is further configured to determine whether to move the first actor based at least in part on a determination of optimization of computing resources within the actor system.

4, The actor system of clause 3, wherein the determination of optimization of computing resources within the actor system is based at least in part on one or more of a global view of the actor system, a regional view of the actor system, or a local view of the actor system.

5, The actor system of clause I, wherein the degree of closeness comprises at least one of binary option or a value in a range of val ses.

6, A method comprising:

monitoring, by a computing device, messages communicated to and from a first actor in a first location of an ac tor system;

determining, by the computing device, a message frequency between the first actor and a second actor in a second location of the actor system;

comparing, by the computing device, the message frequency to a threshold message frequency; and

determining, by the computing device, whether to move the first actor from the fsrst location based on comparison of the message frequency to the threshold message frequency. 7. The method of clause 6. wherein determining whether to move the first actor from the first location comprises:

determining whether the first actor cars be moved to the second location.

8. The method of clause 6, wherein determining whether to move ihe first actor from the first location comprises:

determining that the first actor can be moved to a third location of the actor system; and determining that the first actor would be closer to the second aetor at the third location than at the first location.

9. The method of clause 8, wherein determining that the first actor would be closer to the second actor at the third location than at the first location comprises determining that a rate of messages sent between the first location and the second location is slower than a rate of messages sent between the first location and the third location,

10. The method of clause 6, wherein the threshold message frequency is a predetermi ed threshold frequency.

1 1. The method of clause 6, wherein the threshold message frequency is based on a message frequency between the first aetor and other actors at the first location.

12. The method of clause 6, wherein determining whether to move the first actor from the first location comprises determining that no other locations are available to host the first actor in the actor system.

13. The method of clause 12, further comprising;

sending a signal that additional locations may be provisioned in the actor system.

14. The method of clause 6, wherein the first location comprises a first server and wherein monitoring the messages communicated to and from a first actor in the first location of an actor system Is performed by a management service on the first server. 15. The method of clause 34, wherein the second location comprises a second server and wherein monitoring the messages communicated to and from a first actor in the first location of an actor system comprises determining a message frequency between the first actor and the second server.

16. The method of clause 6, wherein the message frequency s based at feast n part on an overall number of messages sent between die first actor and one or more other actors during a particular period of time, based on sizes of messages sent between the first actor and one or more other actors during a particular period of time, or based on a combination of the number of overall messages and the sizes of messages sent between the first actor and one or more other actors during a particular period of time.

17. A non-transitory computer-readable medium having embodied thereon computer-readable instructions, the computer-readable instructions comprising instructions that, when executed by a computing device, at least cause:

determining to attempt to move a first actor from a first location to a new location within an actor system;

inquiring whether other locations In the actor system are available to host the first actor; determining a number of available locations in the actor system are available to host the first actor and

performing at least one of the following:

leaving the first actor at the first location when the number of available locations is zero,

sending the first actor to an available location when the number of available locations is one, and

determining a destination location from the available locations based on a degree of closeness between the first actor and one or more other actors in the actor system and sending the first actor to the destination location when the number of available locations is greater than one,

I S, The non-transitory computer-readable medium of clause I ?, wherein Inquiring whether other locations in the actor system are available to host the first actor comprises:

sending availability inquiries to a plurality of locations within the actor system. 1 9. The non-transitory computer-readable medium of clause 18, wherein inquiring whether other locations in the actor system are available to host the first actor further comprises:

receiving responses to the availability inquiries (torn the plurality of locations within the actor system.

20. The method of clause 17, wherein determining a number of available locations in the actor system are available to host the first actor is based at least in part on the responses to the availability inquiries.

21. The ησπ-iransitory computer-readable medium of clause ( 7, wherein the computer-readable instructions further comprise instructions that, when executed by the computing device, at least cause:

signaling a possible need to provision additional locations in the actor system when the number of available locations is zero.

22. The non-transitory computer-readable medium of clause 1 7, wherein determining the destination location from the available locations based on a degree of closeness between the first actor and one or mare other actors comprises determining that the destination location is the closest of the available locations to the one or more other actors.

23. The non-transitory computer-readable medium of clause 17, wherein the computer-readable Instructions further comprise instructions that, when executed by the computing device, at least cause:

determining the degree of closeness between the first actor and one or more other actors.

24. The non-transitory computer-readable medium of clause 23, wherein determining the degree of closeness between the first actor and one or more other actors comprises one or more of:

receiving an indication of the degree of closeness from a developer or an operator of the actor system; or

determining the degree of closeness based on messages sent from and received by the fi st actor,

[08 i | Conditional language used herein, such as, among others, "can," "could," "might," "may," "e.g." and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey thai certain embodiments include, while other embodiments do not Include, certain features, eiements and/or steps. Thus, such conditional language is not generally Intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without author input or prompting, whether these features, elements and/or steps are included or are to be performed m any particular embodimen The terms "comprising," 'including," "having" and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations and so forth. Also, the term, "or" is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term "or" means one, some or ail of the elements in the list.

[0092] While certain example embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions disclosed herein, Thus, nothing in the foregoing description is intended to imply that any particular feature, characteristic, step, module, or block is necessary or indispensable.

Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the inventions disclosed herein. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of certain of the inventions disclosed herein.

Claims

WHAT FS CLAIMED IS:

L An actor system comprising;

a plurality of servers in communication with each other;

a plurality of actors hosted by the plurality of servers; and

a management service located on at least a first server of the plurality of servers, the management service configured to;

receive an indicatio of a degree of closeness between a first actor on the first server and a second actor on a second server of the plurality of servers,

send an availability inquiry to one or more of the plurality of servers, receive a response from the one or more of the plurality of servers, determine a destination server from the one or more of the plurality of servers based on a location of the destination server with respect to the second actor, wherein the determination of the destination server is based on the degree of closeness and a location of the destination server, and

move the first actor from the first server to the destination server.

2. The actor system of claim 1 , wherein the destination server is one of the second server or another server located in a server rack in which the second server is located.

3, The actor system of claim I , wherein the management service is further configured to determine whether to move the first actor based at least in part on a determination of optimization of computing resources within the actor system.

4. The actor system of claim 3, wherein the determination of optimization of computing resources within the actor system is based at least in part on one or more of a global view of the actor system, a regional view of tbe actor system, or a local view of the actor system.

5, The actor system of claim l _f wherein the degree of closeness comprises at least one of binary option or a value in a range of values.

6. A method comprising:

monitoring, by a computing device, messages communicated to and from a first actor in a first location of an actor system;

determining, by the computing device, whether to move the first actor from the first location based on comparison of the message frequency to the threshold message frequency.

7. The method of claim 6, wherein determining whether to move the first actor from the first location comprises:

determining that the first actor can be moved to a third location of the actor system; and determining that the first actor would be close to the second actor at the third location than at the first location.

8. The method of claim ?, wherein determining that the first actor would be closer to the second actor at the third location than at the first location comprises determining that a rate of messages sent between the first location and the second location is slower than a rate of messages sent between the first location and the third location.

9. The method of claim 6, wherein the threshold message frequency is a predetermined threshold frequency.

10. The method of clai 6, wherein the threshold message frequency is based on a message frequency between the first actor and other actors at the first location.

1 1. The method of claim 6, wherein determining whether to move the first actor from the first location comprises determining that no other locations are available to host the first actor in the actor system,

12. The method of claim 6, wherein the first location comprises a first server and wherein monitoring the messages communicated to and from a first actor in the first location of an actor system is performed by a management service on the first server. 13, The method of claim 12, wherein Ihe second location comprises a second server and wherem monitoring the messages communicated to and from a first actor in the first location of an actor system comprises determining a message Frequency between the first actor and the second server,

14, The method of claim 6, wherein the message frequency is based at least in part on an overall number of messages sent between the first actor and one or more other actors during a particular period of time, based on sizes of messages sent between the first actor and one or more other actors during a particular period of time, or based on a combination of the number of overall messages and the sizes of messages sent between the first actor and one or more other actors during a particular period of time,

15, A system comprising: one or more processors; and one or more memories having embodied thereon computer-readable instructions, the computer-readable instructions comprising instructions that, when executed by the one or more processors, at least cause:

inquiring whether other locations in the actor system are available to host the first actor; determining & number of available locations in the actor system are available to host the first actor; and

performing at least one of the following:

determining a destination location from the available locations based on a degree of closeness between the first actor and one or more other actors in the actor system and sending the first actor to the destination location when the number of available locations is greater than one.