Cluster Resource Management: A Scalable Approach

1
Cluster Resource Management A Scalable Approach

• Ning Li and Jordan Parker
• CS 736 Class Project

2
Outline

• Introduction
• A Scalable Approach Hierarchy
• Results
• Conclusions
• Questions

3
Why Study Resource Management?

• Clusters have become increasingly popular for
  large parallel computing.
• Web Servers
• Clusters are becoming increasingly large to the
  order of thousands of nodes.
• Clusters are providing multiple services.
• Hard to evaluate
• Bad is easy to determine
• Good is much harder

4
Resource Management Example

• 4th Node Services only B
• Poor Management
• Ideal

Overall
A 37.5
B 62.5
5
Clustering Goals

• Scalability
• Reliability
• High Performance
• Affordability

6
Related Work

• Proportional-Share
• Cluster Reserves

7
Related Work Approach Differences

• Our Goal to provide a scalable solution for
  resource management.
• Other work focused primarily on just having good
  management
• This often meant 1 manager for all the nodes
• Clearly this could present a scalable bottleneck
• Effectiveness Other solutions probably better
  for smaller clusters, we hope to be better for
  large (gt1000 nodes) clusters.

8
Outline

• Introduction
• A Scalable Approach Hierarchy
• Results
• Conclusions
• Questions

9
Hierarchy A Scalable Approach

• Hierarchical Management
• Nodes service jobs
• Managers facilitate resource management

10
Banking Algorithm

• Goal
• Determine best allocation given previous usage
• Primitives
• Tickets
• Bank accounts
• Deposit / withdraw tickets
• 6 Steps

11
Banking Algorithm

• Step 1 For each service class on each node
• Deposit unused tickets
• Step 2 For each service class on each node
• Reallocate service class
• Full utilization Allocation usage k
• Under utilization Allocation usage - k

12
Banking Algorithm Cont.

• Step 3 For each service class
• Compare total allocation to desired
• Subtract from over-allocated
• Add to needy under-allocated
• Step 4 For each service class
• Deposit / Withdraw
• If still over-allocated withdraw
• If still under-allocated deposit

13
Banking Algorithm Cont.

• Step 5
• Withdraw and allocate
• Reward the needy nodes
• Step 6
• Done, clear the bank accounts

14
Reliability

• Bottom-up Manager Replacement

5
6
7
8
9
10
11
12
2
3
4
1
2
5
15
Outline

• Introduction
• A Scalable Approach Hierarchy
• Results
• Conclusions
• Questions

16
Results
Cluster Nodes Managers 1st/2nd Level Reporting 1st/2nd Level Workloads Workloads Class 2 Constraints Tests Tests
4 2/1 1/1 Steady Dyn 1 1 1
1/5 Steady Dyn
100 10/1 1/1 Steady Dyn 1-30 2 3
1/5 Steady Dyn 4 4
900 30/1 1/1 Steady Dyn 1-300 5 5
1/5 Steady Dyn
17
Implementation Details

• Simulations via The NS Network Simulator
• Low bandwidth 10Mbs communication network
• UDP for lower server overhead
• Assumptions
• Node level resource management works ideally

18
Test 1 Overview

• 4 nodes 3 services 60/30/10 Allocation
• 4th node receives all of 3rd classs requests
• Steady Workload

19
Test 1 Data
20
Test 2 Overview

• 100 nodes 3 services 60/30/10 Allocation
• nodes 1-30 receive all of 3rd classs requests
• Steady Workload

21
Test 2 Data
22
Test 3 Overview

• 100 nodes 3 services 60/30/10 Allocation
• nodes 1-30 receive all of 3rd classs requests
• Dynamic Workload

23
Test 3 Data
24
Test 4 Overview

• 100 nodes 3 services 60/30/10 Allocation
• nodes 1-30 receive all of 3rd classs requests
• Steady Workload
• Reporting 1/5
• Nodes every 0.3 second
• Managers every 1.5 seconds

25
Test 4 Data
26
Test 5 Overview

• 900 nodes 3 services 60/30/10 Allocation
• nodes 1-300 receive all of 3rd classs requests
• Steady Workload

27
Test 5 Data
28
Outline

• Introduction
• A Scalable Approach Hierarchy
• Results
• Conclusions
• Questions

29
Conclusions

• Benefits of an hierarchy
• Scalable
• Reliable
• Geographic Applications
• Implemented a new management scheme Banking
• Comparable Results
• Improved Scalability

30
Conclusions

• Clusters are sensitive to small policy changes
• Clusters are built for specific workloads
• Their performance is important and small changes
  have significant impact
• No scheme is universally applicable
• Future Work
• Real system implementation
• Real Workloads
• Real node level resource management
• More steady performance

31
Outline

• Introduction
• A Scalable Approach Hierarchy
• Results
• Conclusions
• Questions

32
Questions
33
Related Work Proportional-Share

• Stride Scheduling
• Ticket based and similar to lottery
• Scale
• Randomly query k nodes to find best allocation
• Different Application
• Condor-like resource allocation/applications

34
Related Work Cluster Reserves

• Resource Container Schedulers
• Constrained Optimization Algorithm
• Scale
• Centralized single manager

35
Hierarchical Cluster Reserves Version 1

• Modify Cluster Reserves optimization algorithm
• Use it when manager manages nodes
• AND when level_n1 manager manages level_n
  managers.

36
Hierarchical Cluster Reserves Version 2

• Cluster Reserves optimization algorithm
• Use it when manager manages nodes
• Dont use it for upper level managers
• Modify the manager to manager reporting
• Lie to the algorithm