/
CSE 591: Energy-Efficient Computing Lecture  15 SCALING:  storage CSE 591: Energy-Efficient Computing Lecture  15 SCALING:  storage

CSE 591: Energy-Efficient Computing Lecture 15 SCALING: storage - PowerPoint Presentation

luanne-stotts
luanne-stotts . @luanne-stotts
Follow
342 views
Uploaded On 2019-11-01

CSE 591: Energy-Efficient Computing Lecture 15 SCALING: storage - PPT Presentation

CSE 591 EnergyEfficient Computing Lecture 15 SCALING storage Anshul Gandhi 347 CS building anshulcsstonybrookedu rabbit paper Objectives Powerproportional storage system PP in the presence of failures ID: 761880

equal policy load data policy equal data load blocks work sierra node storage nodes replicas maintain anshul simple avoid

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "CSE 591: Energy-Efficient Computing Lect..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

CSE 591: Energy-Efficient ComputingLecture 15SCALING: storage Anshul Gandhi 347, CS building anshul@cs.stonybrook.edu

rabbit paper

Objectives Power-proportional storage system PP in the presence of failures Avoid interference between competing apps Avoid writing unnecessary data blocks Maintain only required replicas Considers an HDFS-like FS where all data is important and equally accessed (mostly). This is unlike web data which has skewed popularity.

Simple policy p nodes B/p load (N-p) nodes B/N load (N >> p) primary (r-1) replicas

Requirements of layout policy The equal-work policy ensures equal load sharing . Formally , the equal-work policy is the result of an optimization problem that minimizes p with the constraints, tputi = ( i/p)tputp for all i = p + 1, ..., N for a given replication factor r.

Equal work policy

Equal work policy – LB challengeith node has B/i blocks, but how many requests to these blocks should it serve? i th node hosts more blocks than (i+j) th node.

Fault tolerance

sierra paper

sierra = rabbit + writes

Motivation

Motivation

Challenges Data layout for power savings Maintain read and write availability during failures Predict needed capacity to sustain load

Simple policies

Sierra