1 / 46

Cache Updates in a Peer-to-Peer Network of Mobile Agents

Cache Updates in a Peer-to-Peer Network of Mobile Agents. Elias Leontiadis Vassilios V. Dimakopoulos Evaggelia Pitoura Department of Computer Science University of Ioannina Greece. Introduction. Multi-Agent system (MAS) Network of software agents

marlie
Télécharger la présentation

Cache Updates in a Peer-to-Peer Network of Mobile Agents

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Cache Updates in a Peer-to-Peer Network of Mobile Agents Elias LeontiadisVassilios V. DimakopoulosEvaggelia Pitoura Department of Computer ScienceUniversity of IoanninaGreece

  2. Introduction • Multi-Agent system (MAS) • Network of software agents • Computational resources are distributed across this network • The agents cooperate to fulfill a specified task To do so, they need resources provided by other agents • Open MAS • No global knowledge of the agents in the system • Thus, agents are not aware of which agent provides a particular resource

  3. Resource discovering in open MAS Issue How to locate an agent that provides a particular resource Approaches: • Central directory • There is a single agent which holds a directory, matching resources to agents • Middle agents • Some agents (middle agents) keep a fraction of the directory • Our approach: Distributed caches: each agent keeps part of the directory • Performance • Failure tolerance

  4. Distributed cache model Cache of a1 R4: A2 R8: A6 R1R2R3 • Each agent maintains cache entries about k different resources of other agents • Cache entries: keep the contact information of one agent that provides the resource • The system is modeled as a directed graph G(V;E) called the cache network • If agent u  v, then v is a neighbor of u a1 R4R7 a2 a3 R9 a4 a5 R1 R6 a6 R8R5R2 R1

  5. Searching for a resource • We consider the following flood-based search methods • Plain flood • Forward the query to all its neighbors for a maximum number of steps (TTL: Time-To-Live) • Teeming • Propagates the search message only to a random subset of its neighbors • Φ the fixed probability of selecting a particularneighbor • Teeming with decay • like teeming, but the subset gets smaller as we go deeper into the tree • Φ = (1-d)level d < 1. d is called decay parameter • K-Random paths (K-walkers) • The agent that initiates the search selects K random neighbors • All the other agents forward the message to only one random neighbor Plain flood Teeming 2-Random paths

  6. The Problem: Cache network and mobility • Mobility • Agents may move (e.g. change IP) • Resources may move from one agent to another • Problem: when an agent A moves • Other agents that cached A’s resources before, now have invalid cache entries In fact, no one knows the new location of agent A: when it moved, it didn’t inform anyone New location a3 a1 a2 a3 Old location a4 a5 a6

  7. Cache update policies We consider the problem of cache updates in a peer-to-peer network of mobile agents • We propose a number of update policies that combine two basic techniques: • Pull-based method • Initiated by the agent that wants to update its cache • Push-based method • Initiated by the agent that moves. Since the cache entries form our overlay network, what we update is the network topology itself.

  8. Pull-based method Initiated by the agent that wants to refresh a cache entry (either periodically or on-demand when it discovers that an entry is invalid) • Any flood-based search algorithm can be used • Plain flood • Teeming (with decay) • K-Random paths • Pull method • search the network for an agent that knows the new location

  9. Pull search example Old location Agents knows the new location Agent that pulls New location

  10. Pull search example

  11. Pull search example

  12. Pull search example

  13. Pull search example Replies Directly

  14. Pull search example

  15. Push-based method • Pull cannot work alone When an agent moves, it must inform at least oneother agent about its new location Push method When an agent moves, it “pushes” a message to the network to inform other agents about its new location

  16. Push example Needs the update Old location Needs the update Needs the update New location

  17. Push example

  18. Push example

  19. Push example

  20. Push example

  21. Our Update Policies • Plain Push/Pull combination with appropriate variations of flooding • Push with snooping directories combined with periodic pullsa novel variation of push, where agents that receive information about other moving agents maintain it for a short period of time • Inverted Cache with Leasingan informed push approach combined with leasing

  22. Plain push/pull method • Plain push The moving agent does not know which agents need the update It blindly floods the network with messages that contain its new location. When an agent receives the push message: it updates its cache entries, if needed

  23. Plain push/pull method • A wide push is needed to inform a sufficient number of agents • Large TTL and small decay • An agent may not receive the update because: • Offline during push • Push messages may not reach it • Larger TTL and decay values needed • Disconnected network • Such agents should perform on-demand pull.

  24. Push with snooping directories and periodic pulling Push with snooping directories Every agentmonitors the network and maintains a directory of recently moved agents. This directory is termed snooping directory. • How it works: • Anagent that receives a push message concerning other moving agentsstores this information in its own snooping directory • It does so, even if it doesn’t need this information to update its own cache. • Keeps it for a small period of time (expiration time) So, each agent remembers the new location of every recently moved agent that came to its knowledge

  25. Push with snooping directories example

  26. Push with snooping directories example Snooping dir -a1 moved to … Snooping dir -a1 moved to …

  27. Push with snooping directories example Snooping dir -a1 moved to … Snooping dir -a1 moved to … Snooping dir -a1 moved to … Snooping dir -a1 moved to … Snooping dir -a1 moved to … a1

  28. Push with snooping directories and periodic pulling • All agents perform periodic pulls Periodically, they search the network for agents in their cache that have recently moved and update the cache, if necessary • Why periodic pulling? • To take advantage of the snooping directories: if we pull after a long time, information about old moves might have already been deleted • Time between two subsequent pulls < expiration time of entries in the snooping directories

  29. Push with snooping directories and periodic pulling:Discussion • This method allows the use of narrower push/pull flooding Less message overhead For example: If agents pull periodically from their two-hop neighborhood: • All the nodes that are two-hop away from push-informed agents will eventually receive the update when they pull. • So, it is sufficient to push-inform just one agent in each two-hop neighborhood • We prefer to use a k-walkers algorithm for pushing • We spawn K-walkers and we require that all agents be two-hops away from the walk-paths at the most

  30. Push with snooping directories Changed location

  31. Inverted cache push/pull • Every agent keeps a list of the agents to which it is known, called inverted cache. • When the agent moves, it informs the agents found • in its inverted cache to update their plain cache • in its plain cache to update their inverted directory • By knowing where to send the updates • Avoid flooding  Low message overhead • Drawback: When an agent adds/replaces/deletes a resource from its cache, a message has to be sent to the resource owner. A B C A C B

  32. Inverted cache push/pull Issue Storing the entire inverted cache directory may not always be preferable, as there may exist popular agents/resources Solution • Only a limited directory may be maintained • The inverted cache strategy can be combined with leasing • The agents that are not informed could use on-demand pull.

  33. Inverted cache push/pull: Leasing • Every entry in the cache gets a lease time issued by the resource owner • After the expiration the resource owner may delete the entry from its inverted cache, without ever informing the leaser • We could control the size of inverted cache directory through lease times • Shorter lease times  smaller inverted cache directories Lease time Time interval, during which the resource owner guarantees that it will notify the leaser in case the former moves.

  34. Experimental results • Evaluation using a simulator • Create a random graph • Each agent shares some resources • Some resources (few) are more popular than others • Initially, all agents have valid cache entries • The simulation runs for a number of turns: • At each turn, an agent can • Move • Search for a resource (on demand pull if necessary) • Make a cache replacement etc. • We keep statistics • Push/pull messages • Percentage of valid cache entries • Average directory sizes • Number of steps needed for the update to propagate

  35. Simulation • There are 1000 agents owning 3000 resources. We run the simulation for 250 turns. • We are mainly interested in: • The percentage of valid cache entries during the simulation • The message overhead produced by • Pull • Push • Cache replacements

  36. Plain push/pull Extend of flooding and percentage of valid cache entries

  37. Plain push/pullExtend of flooding and message overhead

  38. Push with snooping directories and periodic pullExtend of flooding and percentage of valid cache entries

  39. Push with snooping directoriesand periodic pullExtend of flooding and message overhead

  40. Plain push/pull vs. snooping

  41. Inverted cache push/pullwith leasesLease time duration and percentage of valid cache entries

  42. Inverted cache push/pullReplace frequency and message overhead

  43. Comparison • Plain push/pull • Should be used only when we do not want to use any additional memoryfor cache update methods • Can be used in an unreliable environment (on-demand pull) • Achieves satisfactory results when using wide flooding • High message overhead: It should be avoided when we have high mobility • Push with snooping directories and periodic pull • Achieves the same cache quality with plain push/pull but with significantly less message overhead. • Uses additional memory • Inverted cache with leasing • Negligible push message overhead • One-hop update propagation • Replacing a cache entry requires contacting the resource owner • Unsuitable for systems with high replacement ratio. • Directory size can become quite large for popular agents • Not appropriate for unreliable open MAS • Agents rely on each other to be updated • Agents must be online to maintain a valid inverted cache

  44. Conclusions • We considered the problem of cache updates in a peer-to-peer network of mobile agents • Each agent maintains in its cache information about other agents • When agents move, cached entries about them become obsolete • We propose a number of update policies that combine • pull-based techniques, that are initiated by the agent that wants to update its cache • push-based methods, that are initiated by the agent that moves. • Push/pull variations • We propose a novel variation of push, where agents that receive information about other moving agents maintain it for a short period of time in a snooping directory • We propose an Informed push approach and we combine it with leasing (inverted cache) • Our experimental results designate • Snooping directory leads to the attainment of the same cache consistency compared with plain push/pull but with ten times less message overhead • Inverted cache method is message-cost effective but only when cache replacements are not too frequent

  45. Future work • Mobile agents resemble mobile (wireless) nodes in ad-hoc networks: apply our policies to message routing and resource discovery in these networks • File replicas in p2p systems apply our policies to keep replicas consistent

  46. Thank you • More details • www.cs.uoi.gr/~ilias/dcache/

More Related