© 2003, Carla Ellis Self-Scaling Benchmarks Peter Chen and David Patterson, A New Approach to I/O Performance Evaluation – Self-Scaling I/O Benchmarks, Predicted I/O Performance, SIGMETRICS 1993.

Workloads Experimental environment prototype real sys exec- driven sim trace- driven sim stochastic sim Live workload Benchmark applications Micro- benchmark programs Synthetic benchmark programs Traces Distributions & other statistics monitor analysis generator Synthetic traces “Real” workloads Made-up © 2003, Carla Ellis Data sets You are here

© 2003, Carla Ellis Goals A benchmark that automatically scales across current and future systems –It dynamically adjusts to system under test Predicted performance based on self- scaling evaluation results –Estimate performance for unmeasured workloads –Basis for comparing different systems

© 2003, Carla Ellis Characteristics of an Ideal I/O Benchmark Benchmark should 1.Help in understanding why, isolate reasons for poor performance 2.Be I/O limited 3.Scale gracefully 4.Allow fair comparisons among machines 5.Be relevant to a wide range of applications 6.Be tightly specified, reproducible, explicitly state assumptions Current benchmarks fail

© 2003, Carla Ellis Overview of Approach Step 1: scaling: Benchmark automatically explores workload space to find relevant workload. –By depending on system under test, the ability to compare systems on benchmark results is lost Step 2: Predicted performance scheme helps restore that capability –Accuracy of prediction must be assured

© 2003, Carla Ellis Workload Parameters uniqueBytes – total size of data accessed sizeMean – average size of an I/O request –Individual requests chosen from normal distribution readFrac – fraction of reads; fraction of writes is 1- readFrac seqFrac – fraction of requests that are sequential access –For multiple processes, each has its own thread processNum – concurrency Workload is user-level program with parameters set

Representativeness Does such a synthetic workload have the “right” set of parameters to capture a real application (characterized by its values for that set of parameters)?

Benchmarking Results Set of performance graphs, one for each parameter, while holding all other parameters fixed at their focal point values. –75% performance point –Found by iterative search process More of workload space is explored Does not capture dependencies among parameters

focal point = (21MB, 10KB, 0,1,0)

© 2003, Carla Ellis Families of Graphs General applicability – representative across range of parameter (75% rationale) Multiple performance regions – especially evident for uniqueBytes because of storage hierarchy issues –On border – unstable –mid-range focal points

cache disk Larger requests better Reads are better than writes Sequential helps Sequential has little effect

© 2003, Carla Ellis Predicted Performance Problem: benchmark chosen will be different for 2 different systems so they can not be directly compared. Solution: Estimate performance for unmeasured workloads so a common set of benchmarks can be used for comparisons

© 2003, Carla Ellis How to Predict Assume the shape of performance curve for one parameter is independent of values of other parameters. Use self-scaling benchmark to measure with all but one parameter fixed at focal point

Solid lines measured performance with sizeMean fixed on left (S f ), processNum fixed on right (P f ) Predict throughput curve with sizeMean at S 1 by assuming constant ratio Throughput(processNum, sizeMean f ) Throughput(processNum, sizeMean 1 ) which is known at processNum P f in righthand graph

Accuracy of Predictions For SPARCstation + 1disk Measured at random points in parameter space. Error correlated to uniqueBytes

Comparisons

For Discussion Next Thursday (because of snow) Survey the types of workloads – especially the standard benchmarks – used in your proceedings (10 papers). is a great resource © 2003, Carla Ellis

Continued discussion of reinterpreting an experimental paper into strong inference model