Presentation is loading. Please wait.

Presentation is loading. Please wait.

Distillation of Performance- Related Characteristics

Similar presentations


Presentation on theme: "Distillation of Performance- Related Characteristics"— Presentation transcript:

1 Distillation of Performance- Related Characteristics kurmasz@cc.gatech.edu

2 Introduction zWant synthetic workload to maintain certain realistic properties or attributes yWant representative behavior (performance) zResearch Question: yHow do we identify needed attributes? zWe have a method...

3 Goal CDF of Response Time will have performance similar to original. (R,1024,120932,124) (W,8192,120834,126) (W,8192,120844,127) (R,2048,334321,131... Original Workload Given a workload and storage system, automatically find a set of attributes, so Attribute List SyntheticWorkload (R,1024,120932,124) (W,8192,120834,126) (W,8192,120844,127) (R,2048,334321,131... synthetic workloads with the same values

4 Why? zPredicting performance of complex disk arrays is extremely difficult. yMany unknown interactions to account for. zList of attributes much easier to analyze than large, bulky workload trace. zList of attributes tells us: yWhich patterns in a workload affect performance yHow those patterns affect performance zPossible uses of attribute lists: yOne possible basis of “similarity” for workloads yStarting point for performance prediction model

5 Basic Idea zAttribute list may be different for every workload/storage system pair yRequire general method of finding attributes yMust require little human intervention zBasic Idea: Add attributes until performance of original and synthetic workloads is similar. (R,1024,120932,124) (W,8192,120834,126) (W,8192,120844,127) (R,2048,334321,131... Original Workload Attribute List SyntheticWorkload (R,1024,120932,124) (W,8192,120834,126) (W,8192,120844,127) (R,2048,334321,131...

6 Mean Arrival Time Arrival Time Dist. Hurst Parameter Mean Request Size Request Size Dist. Request Size Attrib 3 Request Size Attrib 4COV of Arrival Time Dist. of LocationsRead/Write ratio Mean run length Markov Read/Write Jump DistanceR/W Attrib. #3 Proximity MungeR/W Attrib #4 Mean Read Size D. of (R,W) Locations Read Rqst. Size Dist.Mean R,W run length Mean (R, W) SizesR/W Jump Distance (R, W) Size Dists.R/WProximity Munge Mean Arrival Time Arrival Time Dist. Hurst Parameter Mean Request Size Request Size Dist. Request Size Attrib 3 Request Size Attrib 4COV of Arrival Time Dist. of LocationsRead/Write ratio Mean run length Markov Read/Write Jump DistanceR/W Attrib. #3 Proximity MungeR/W Attrib #4 Mean Read Size D. of (R,W) Locations Read Rqst. Size Dist.Mean R,W run length Mean (R, W) SizesR/W Jump Distance (R, W) Size Dists.R/WProximity Munge Choosing Attribute Wisely zProblem: yNot all attributes useful yCan’t test all attributes zOur Solution: yGroup attributes yEvaluate entire groups at once Attributes How are they grouped? How are they evaluated?

7 zWorkload is series of requests y(Read/Write Type, Size, Location, Interarrival Time) zAttributes measure one or more parameters yMean Request Size Request Size yDistribution of LocationLocation yBurstinessInterarrival Time yRequest Size yRead/Write zAttributes grouped by parameter(s) measured yLocation = {mean location, distribution of location, locality, mean jump distance, mean run length,...} Attribute Group Distribution of Read Size

8 Evaluate Attribute Group zAdd “every” attribute in group at once and observe change in performance. zAmount of change in performance estimator of most effective attribute “All” (Size, R/W) “All” Request Size “All” Location

9 “All” Location attribute “All” (Location, Request Size) attribute The “All” Attribute zThe list of values for a set of parameters contains every attribute in that group zAttributes in that group will have same value for both original and synthetic workload z List represents “perfect knowledge” of group

10 RMS/Mean : Original:.1877 Current:.0918

11 Main Ideas zNew method of automatically finding performance-related attributes: yMeasure completeness of list by comparing performance of synthetic workloads yUseful method of grouping attributes yEffective method of evaluating entire groups of attributes yAvoid evaluation of useless attributes zkurmasz@cc.gatech.edu zwww.cc.gatech.edu/~kurmasz

12 END OF SHORT TALK zThe rest of the slides are for the full talk. zCurrent 26 January, 6:44 pm

13 Implement Improvement zAdd attribute from chosen group zThis is most time-consuming part yOnly a few attributes known, so we must develop most attributes from scratch zThis should get easier as technique used and “attribute library” grows yFuture Work: We will eventually need an intelligent method of searching library

14 Main Research Focus z1). How to automatically choose and apply “additive” or “subtractive” method z2). How to automatically evaluate results and choose single attribute group zIn practice, there are subtleties that are easily addressed by hand, but difficult to generalize for algorithm.

15 Current Progress zWe have working application yAmbiguous cases still done by hand yApplication stops and asks for a hint yAlgorithm being improved incrementally so that it needs fewer hints zApplication used on Open Mail Workload

16 The “All” Attribute zThe list of values for a set of parameters contains every attribute in that group yAttributes in that group will have same value for both original and synthetic workload “All” attribute for location


Download ppt "Distillation of Performance- Related Characteristics"

Similar presentations


Ads by Google