1 / 45

Armend Hoxha

Mizan : A system for Dynamic Load Balancing in Large-Scale Graph Processing. Presented by:. Armend Hoxha. Trevor Hodde. Kexin Shi. A System for Dynamic Load Balancing. Mizan ( a rabic ) : a double-pan scale. Pregel HADI Pegasus X-RIME.

rupert
Download Presentation

Armend Hoxha

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. Mizan: A system for Dynamic Load Balancing in Large-Scale Graph Processing Presented by: Armend Hoxha Trevor Hodde Kexin Shi

  2. A System for Dynamic Load Balancing Mizan(arabic) : a double-pan scale • Pregel • HADI • Pegasus • X-RIME We’ll focus on Pregel, since Mizan is a refined Pregel system.

  3. Pregel • A scalable graph mining system • Message passing based programming model • It improved the overall performance by 1-2 orders of magnitude over a traditional MapReduce implementation for a • wide array of graph mining algorithms: • Page Rank • Shortest paths problems • Bipartite matching • Semi-clustering

  4. Pregel • A scalable graph mining system • Message passing based programming model • It improved the overall performance by 1-2 orders of magnitude over a traditional MapReduce implementation for a • wide array of graph mining algorithms: • Page Rank • Shortest paths problems • Bipartite matching • Semi-clustering

  5. Pregel • A scalable graph mining system • Message passing based programming model • It improved the overall performance by 1-2 orders of magnitude over a traditional MapReduce implementation for a • wide array of graph mining algorithms: • Page Rank • Shortest paths problems • Bipartite matching • Semi-clustering • Existing implementations focus primarily • on graph partitioning as a preprocessing • step to balance computation.

  6. Pregel is built on BSP programming model BSP = Bulk Synchronous Parallel This picture is taken from the paper

  7. Pregel is built on BSP programming model • Each vertex runs an algorithm and can send messages asynchronously to any other vertex • During each superstep vertices run in parallel across a distributed infrastructure • Each vertex processes incoming messages from the previous superstep • A vertex is said to be active if it is processing and/or sending messages to other vertices • This process continues until all vertices have no messages to send – they all become inactive This picture is taken from the paper

  8. Graph Algorithms • Stationary • An algorithm is stationary if its active vertices send and receive the same distribution of messages across supersteps. At the and of each superstep, all active nodes become inactive. • Examples: Page Rank, Diameter Estimation, Finding Weakly Connected Components etc. • Non-Stationary • An algorithm is non-stationary if the size of its outgoing messages changes across supersteps. Such variations can create workload imbalances across supersteps. • Examples: Distributed Minimal Spanning Tree Construction (DMST), Graph Queries, various simulations on social network graphs – advertisement propagation etc.

  9. Page Rank (stationary) against DMST (non-stationary) This picture is taken from the paper Difference between stationary and non-stationary algorithms. Total represents the sum across all workers and Max represents the maximum amount on a single worker. Both of these algorithms were run on a cluster of 21 machines and processed the same dataset (LiverJournal1). The input graph was partitioned using a hash function.

  10. What causes imbalance in non-stationary algorithms? • There are two sources of imbalance: • one originating from graph structure, and • another from the algorithm behavior This picture is taken from the paper

  11. Graph partitioning • Common approaches to partitioning the data are: • Hash-based • Range-based • Min-cuts Divide the dataset based on a simple heuristic: to evenly distribute vertices across compute nodes, irrespective of their edge connectivity. Considers the vertex connectivity and partitions the data such that it places strongly connected vertices close to each other (on the same cluster).

  12. Graph partitioning • Common approaches to partitioning the data are: • Hash-based • Range-based • Min-cuts Divide the dataset based on a simple heuristic: to evenly distribute vertices across compute nodes, irrespective of their edge connectivity. Considers the vertex connectivity and partitions the data such that it places strongly connected vertices close to each other (on the same cluster). Both of these partitioning strategies result in graph dependent performance.

  13. Graph partitioning Because of its size, arabic-2005 couldn’t be partitioned using min-cuts in the local cluster where experiments were conducted. Run Time (Min)

  14. Balanced Computation and Communication Balancing the computation and communication is fundamental to the efficiency of a Pregel System Existing implementations of Pregel take one or more of the following five approaches to achieving a balanced workload: Provide simple graph partitioning schemes, like hash- or range-based partitioning (e.g. Giraph) Allow developers to set their own partitioning scheme or pre-partition the graph data (e.g. Pregel) Provide more sophisticated partitioning techniques (e.g. GraphLab, GoldenOrb and Surfer use min-cuts) Utilize the distributed data stores and graph indexing on vertices and edges (e.g. GoldenOrb and Hama), and Perform coarse-grained load balancing (e.g. Pregel)

  15. Balanced Computation and Communication Balancing the computation and communication is fundamental to the efficiency of a Pregel System Existing implementations of Pregel take one or more of the following five approaches to achieving a balanced workload: Provide simple graph partitioning schemes, like hash- or range-based partitioning (e.g. Giraph) Allow developers to set their own partitioning scheme or pre-partition the graph data (e.g. Pregel) Provide more sophisticated partitioning techniques (e.g. GraphLab, GoldenOrb and Surfer use min-cuts) Utilize the distributed data stores and graph indexing on vertices and edges (e.g. GoldenOrb and Hama), and Perform coarse-grained load balancing (e.g. Pregel) • All of the approaches above assume that: • The structure of the graph is static, • The algorithm has predictable behavior or • Developer knows about the runtime characteristics of the algorithm.

  16. New Approaches Are Needed • From the experiments presented before, we saw that: • When graph mining algorithm has unpredictable communication needs, • Frequently changes the structure of the graph, or • Has a variable computation complexity • A single solution is no enough. • We want to build a system that is: • Adaptive • Agnostic to the graph structure, and • Requires no a priori knowledge of the behavior of the algorithm

  17. M I Z A N

  18. What is Mizan? Binary Space Partitioning graph processing framework A way of “recursively subdividing space” into sets which are collected into a graph (or in this case, a tree) Open source, written in C++ Extends the Pregel API (you can read more here) Pregel is a scalable system for “Large-Scale Graph Processing” Known to improve MapReduce performance by 1-2 orders of magnitude Focuses on efficient load balancing across workers in a cluster Two major steps in this process Monitoring Migration Planning

  19. Monitoring (The easy part) • Mizan monitors certain metrics for each worker node to ensure equal load balancing • Number of outgoing messages to other nodes • Only messages sent to remote workers are counted, not messages rerouted to the same worker • Total incoming messages • This includes messages rerouted to the same worker because queue size can affect performance • Response time • Measured from the time at which the node begins processing a message until it finishes

  20. Migration Planning • Objectives: • Decentralized • Make sure all processing happens in parallel to improve performance • “Simple” <- The writers words…not mine • Transparent • Extends the Pregel API but does not need to change it • Does not assume anything about a graph structure or algorithm

  21. Migration Planning (more) • 5 easy steps to find the “strongest cause” of workload imbalance among the three monitored metrics • Provides a way to detect and (hopefully) prevent instability due to workload imbalance

  22. Migration Planning (the steps) • Step 1: Identify the source of imbalance • Compare the worker’s execution time against a normal distribution and flag any outliers • Remember: Mizan monitors each vertex for those 3 metrics • Outgoing Messages to remote nodes • All incoming messages • Response time

  23. Migration Planning (step 1 picture) This picture is taken from the paper

  24. Migration Planning (the steps) • Step 1: Identify the source of imbalance • Step 2: Select the migration objective • Detects the largest cause of workload imbalance by comparing metrics for incoming and outgoing messages for each worker with the worker’s execution time • The result of this comparison will tell the system what needs optimization: • Optimize outgoing messages • Optimize incoming messages • Optimize response time

  25. Migration Planning (the steps) • Step 1: Identify the source of imbalance • Step 2: Select the migration objective • Step 3: Pair over-utilized workers with under-utilized ones • All workers create and execute the migration plan in parallel • Each worker can be paired with up to one other worker This picture is taken from the paper

  26. Migration Planning (the steps) • Step 1: Identify the source of imbalance • Step 2: Select the migration objective • Step 3: Pair over-utilized workers with under-utilized ones • Step 4: Select vertices to migrate • This is based on the migration objective determined by the previous steps

  27. Migration Planning (step 4 explained) • Example: • Migration Objective: balance outgoing messages • Solution: select vertices with a large number of outgoing messages and migrate to an underutilized worker

  28. Migration Planning (the steps) • Step 1: Identify the source of imbalance • Step 2: Select the migration objective • Step 3: Pair over-utilized workers with under-utilized ones • Step 4: Select vertices to migrate • Step 5: Migrate the vertices • Challenges: • Migrating vertices across workers while maintaining vertex ownership and fast updates • Minimizing migration costs to ensure that a migration actually helps reduce workload

  29. Maintaining Vertex Ownership • Mizan uses a distributed hash table to implement a lookup service • Vertex v can execute at any worker • V’s home worker ID = hash(ID) mod N • Workers ask the home worker of v for its location • The home worker is notified with the new location as v migrates

  30. Maintaining Vertex Ownership This picture is taken from the paper

  31. Vertices with Large Message Size • Delayed Migration is possible for vertices with very large message sizes • During the migration phase, only the ownership of vertex v is moved to its new worker • Then, at a later time, the actual vertex is passed to the new worker

  32. Vertices with Large Message Size This picture is taken from the paper

  33. Vertices with Large Message Size This picture is taken from the paper

  34. Vertices with Large Message Size This picture is taken from the paper

  35. Experimental Setup • Mizan is implemented on C++ with MPI with three variations: • Static Mizan: Emulates Giraph, disables dynamic migration • Work stealing (WS): Emulates Pregel'scoarse-grained dynamic load balancing • Mizan: Activates dynamic migration

  36. Experimental Setup • Runs on local cluster of 21 machines: • Mix of i5 and i7 processors with 16GB RAM Each • IBM Blue Gene/P supercomputer with 1024 compute nodes: • Each is a 4 core PowerPC450 CPU at 850MHz with 4GB RAM

  37. Dataset • Experiments on public datasets: • Stanford Network Analysis Project (SNAP) • The Laboratory for Web Algorithmics (LAW) • Kroneckergenerator This picture is taken from the paper

  38. Giraph vs. Static Mizan Both pictures are taken from the paper

  39. Effectiveness of Dynamic Vertex Migration • PageRank on a social graph (LiveJournal1) • The shaded columns: algorithm runtime • The unshaded columns: initial partitioning cost This picture is taken from the paper

  40. Effectiveness of Dynamic Vertex Migration • Comparing Work stealing (Pregel clone) vs. Mizan using DMST and ad propagation simulation, on a metis partitioned social graph (LiveJournal1) This picture is taken from the paper

  41. Scalability of Mizan This picture is taken from the paper

  42. Scalability of Mizan This picture is taken from the paper

  43. Conclusion • Mizan is a Pregel system that uses fine-grained vertex migration to load balance computation and communication across supersteps • Mizan is an open source project developed within InfoCloudgroup in KAUST in collaboration with IBM, programmed in C++ with MPI • Mizan scales up to thousands of workers • Mizan improves the overall computation cost between 40% up to two orders of magnitude with less than 10% migration overhead

  44. References • http://www.cs.cornell.edu/~djwill/pubs/mizan.pdf • http://kowshik.github.io/JPregel/pregel_paper.pdf • http://thegraphsblog.files.wordpress.com/2012/11/z-presentation.pdf

  45. Thank you!

More Related