Coordinated Workload Scheduling

1
Coordinated Workload Scheduling

• A New Application Domain for Mechanism Design
• Elie Krevat

2
Introduction

• Distributed systems becoming larger, more complex
• Nodes perform computation and storage tasks
• Workloads enter system and are distributed across
  nodes
• Clients run many workloads, can pay for resources
• Nodes service many workloads (not dedicated)
• System provides QoS guarantees
• Performance load balance workloads to faster
  free nodes
• Efficiency minimize cycles wasted when tasks
  available
• Fairness nodes share resources across workloads

3
Benefits of Shared Storage
Why cluster? Scaling, cost, and management. Why
share? Slack sharing, economies of scale,
uniformity.
4
Throughput Performance Insulation in Shared
Storage

• Each of n workloads on a server
• Executes efficiently within its portion of time
  (timeslice)
• Ideally gets 1/n of its standalone performance
• In practice within a fraction of the ideal
• Argon project Wachs07 provides bounds on
  efficiency across workloads for one server
• Problems extending to many servers
  (cluster-style)
• Synchronized workloads need coordination of
  schedules
• Performance of system limited by slowest node

5
Timeslice challenges
140 ms
Server A
Workload 1
Workload 2
Workload 3
Workload 1
Workload 2
280 ms
Server B
Workload 4
Workload 1
Workload 4
100 ms
Server C
1
6
3
5
2
3
6
1
6
Cluster-style Storage Systems
Data Block
Synchronized Read
1
R
R
R
R
2
3
Client
Switch
Data Fragment
1
2
3
4
4
Client now sends next batch of requests
Storage Servers
6
7
Environment Assumptions

• One client per workload
• Bounded number W of workloads, N of nodes
• Constant set of workloads to be scheduled
• But mechanism might support changing set
• Communication doesnt interfere with
  computation/storage tasks

8
Workload Distribution Settings

• Two alternative workload distribution settings
• Setting I Free Workload Assignment
• Workloads can be freely assigned to many nodes
• Example Embarrassingly parallel distributed apps
• Problem Determine best set of nodes to assign
• Setting II Fixed Workload Assignment
• Workloads must be assigned to fixed set of nodes
• Example Cluster-style storage
• Problem Coordinate responses of nodes with
  better timeslice scheduling

9
Computing Environments with Monetary Incentives

• Workloads pay for resources
• Weather forecasting
• Seismic measurement simulations of oil fields
• Distributed systems sell resources
• Supercomputing centers sell resources
• Shared infrastructures
• Grid computing
• Individually-owned computers sell spare cycles
• SETI_at_Home for
• May not have single administrative domain

10
Why Mechanism Design?

• Central coordinator(s) lack per-node information
• Different performance capabilities and revenue
  models
• Enforce cooperation and global QoS
• Efficiency and fairness not always goals of
  players
• Reduce scheduling problems to general mechanism
• Scheduling coordinated workloads is hard (proof
  later)
• Divide scheduling problems across nodes
• Design mechanism to produce coordination

11
Outline

• Background and Motivation
• ?Mechanism I Free Workload Assignment
• Mechanism II Fixed Workload Assignment
• Conclusions

12
Revenue Model Free Assignment

• Clients pay nodes directly after task
• Payment is per-workload
• Amount depends on many factors
• Speed of response
• Number of requests/computations per timeslot
• Clients may also pay fixed cost to central
  scheduler
• Workloads want the best and fastest nodes
• Central scheduler doesnt know load/speed of
  nodes
• Nodes are greedy and want lots of workloads
• May lie about load/speed if asked directly
• System Goal Assign workloads to nodes that will
  respond fastest

13
Mechanism Design VCG

• Run auction to decide which M nodes to assign
• FIFO approach to scheduling each workload
• Can also run combinatorial auction on bundles
• Nodes respond with bids
• Valuations depend on speed and current load
• Same factors that affect final payment
• Apply Vickrey-Clarke-Groves mechanism
• First auction iteration finds top M bids
• Remove Node X, recompute top M bids
• Additional auction iteration not actually
  necessary
• Difference between Xs bid and M1st bid is
  payment
• May also normalize payments to share wealth over
  nodes

14
Mechanism Results

• Incentive compatible
• Nodes have no incentive to lie, since if they
  over-report valuation for workload theyll still
  be paid true valuation
• Global efficiency (i.e., best allocation for
  workload)
• Related to general task allocation problem
  Nisan99
• k tasks allocated to n agents
• Goal is to minimize completion time of last
  assignment (make-span)
• Valuation of agent is negation of total time
  spent on tasks
• Approximation/randomized algorithms exist for CA

15
Outline

• Background and Motivation
• Mechanism I Free Workload Assignment
• ?Mechanism II Fixed Workload Assignment
• Conclusions

16
Revenue Model Fixed Assignment

• Nodes paid by system at every timestep
• Payment is part of mechanism payment scheme
• System wants quick resolution of workload
  requests
• Nodes need monetary incentives to schedule fully
  coordinated workloads efficiently and fairly
• All M nodes service workload in same timeslice
• Uncoordinated workloads not important
• System Goals
• Enforce coordination of workloads per timeslice
• Achieve fair distribution of resources
• Achieve efficient schedule allocations

17
Coordination is hard
1
2
3
4

• Reduce Max Independent Set problem to problem of
  scheduling max of fully coordinated workloads
  per timeslice
• For every Node xi that services a workload wi,
  then wi has a dependency edge to all other
  workloads serviced by xi
• NP-Complete, but approximation algorithms exist
• For above example, max independent set is 1,3

18
Properties of Schedule Allocations

• Two types of schedule allocations
• Basic quanta timeslice allocation a
• Longer sequence of timeslices atot
• Set of workloads serviced by node nx is Sx
• / timeslice quanta allocated per workload wi is
  qi
• Total quanta count Qx for each node nx
• Delay between consecutive workload schedules in
  allocation atot is schedule distance di,k
• k refers to schedule instance in atot
• Average schedule distance di,avg, per-node is
  dx,avg
• Maximum schedule distance di,max, per-node is
  dx,max

19
Formulas for Schedule Allocation Properties
20
Possible Payment Scheme

• Node is paid max of P credits for each scheduled
  time quanta
• No credits for uncoordinated schedule
• For every cycle of time that workload isnt
  scheduled, payment decreases by c (c ltlt P)
• Node is fined F if starves workload over a period
  of quanta greater than Qthr
• Using derived properties of schedule allocations,
  each node calculates payments

21
Mechanism Design Open Research Problem

• Goal is to improve efficiency and fairness
• Nodes compute their best allocations (through
  heuristics) using payment scheme that rewards
  efficiency/fairness
• Send valuations to central scheduler
• General mechanism determines best global
  allocation
• But coordination is hard optimization problem
• May be better suited only for central scheduler
• Expected properties of a mechanism
• Nodes are players
• No additional utility past payments?
• Auctioned good may be single or total allocations
• Tradeoff is ability to adapt to changing
  workloads vs. better assessment of efficient
  allocations over longer time

22
Outline

• Background and Motivation
• Mechanism I Free Workload Assignment
• Mechanism II Fixed Workload Assignment
• ?Conclusions

23
Conclusions

• Distributed systems environments provide new
  applications for mechanism design
• Goals of better global performance, efficiency,
  fairness
• Not always shared by individual nodes
• Model and analysis of 2 different distribution
  settings
• Free workload assignment solved with VCG
• Fixed workload assignment still open problem
• Revenue model and goals of mechanism vary
• Payment functions use derived allocation
  properties
• Coordination of workloads is hard optimization
  problem
• Motivation for further research in related areas