Upload
ksnarooka
View
228
Download
0
Embed Size (px)
Citation preview
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 1/86
1
UC Berkeley
* ,Director IntelResearchB e rke le y
:// . . .h ttp a b ov e th e clou d s cs b e rkele y/e d u
Cloud Computing:Past, Present, and Future
Professor Anthony D. Joseph*, UC BerkeleyReliable Adaptive Distributed Systems Lab
RWTH Aachen22 March 2010
UC Berkeley
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 2/86
RAD Lab 5-year Mission
Enable 1 person to develop, deploy, operate next -generation Internet application
• Key enabling technology: Statistical machine learning – debugging, monitoring, pwr mgmt, auto-configuration, perf
prediction, ...
• Highly interdisciplinary faculty & students – PI’s: Patterson/Fox/Katz (systems/networks), Jordan
(machine learning), Stoica (networks & P2P), Joseph(security), Shenker (networks), Franklin (DB)
– 2 postdocs, ~30 PhD students, ~6 undergrads
• Grad/Undergrad teaching integrated with research
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 3/86
Course Timeline
• Friday – 10:00-12:00 History of Cloud Computing:
Time-sharing, virtual machines,datacenter architectures, utility computing
– 12:00-13:30 Lunch – 13:30-15:00 Modern Cloud Computing:
economics, elasticity, failures – 15:00-15:30 Break – 15:30-17:00 Cloud Computing
Infrastructure: networking, storage,computation models
• Monday – 10:00-12:00 Cloud Computing research
topics: scheduling, multiple datacenters,testbeds
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 4/86
NEXUS: A COMMONSUBSTRATE FOR CLUSTERCOMPUTING
, , ,Joint work with Benjamin Hindman Andy Konwinski Matei Zaharia Ali,Ghodsi
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 5/86
Recall: Hadoop on HDFS
datanode daemon
Linux file system
…
tasktracker
slave node
datanode daemon
Linux file system
…
tasktracker
slave node
datanode daemon
Linux file system
…
tasktracker
slave node
namenode
namenode daemon
job submission node
jobtracker
, , , & - ,A d a p te d fro m slid e s by Jim m y Lin C h risto p h e B iscig lia A a ron K im b all S ie rra M ich e ls S le ttve t G oo g le D istrib u te d
, ( . )C om p utin g S em ina r 2 0 0 7 licen sed u nd er C rea tion C om m on s Attribu tion 3 0 Licen se
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 6/86
Problem
• Rapid innovation in cluster computingframeworks
• No single framework optimal for all
applications• Energy efficiency means maximizing
cluster utilization
• Want to run multiple frameworks in asingle cluster
•
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 7/86
What do we want to run in thecluster?
Dryad
ApacheHama
Pregel
Pig
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 8/86
Why share the cluster betweenframeworks?
• Better utilization and efficiency (e.g.,take advantage of diurnal patterns)
•
• Better data sharing acrossframeworks and applications
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 9/86
Solution
Nexus is an “operating system” for thecluster over which diverse frameworkscan run
– Nexus multiplexes resources betweenframeworks
– Frameworks control job execution
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 10/86
Goals
• Scalable
• Robust (i.e., simple enough toharden)
• Flexible enough for a variety of different cluster frameworks
• Extensible enough to encourageinnovative future frameworks
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 11/86
Question 1: Granularity of Sharing
Option: Coarse-grained sharing – Give framework a (slice of) machine for its
entire duration
–
Hadoop 1
Hadoop 2
Hadoop 3
ata localitycompromised if
machine held for longtime
Hard to account for new frameworks and changing->demands urts
tilization andinteractivity
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 12/86
Nexus: Fine-grained sharing – Support frameworks that use smaller tasks
(in time and space) by multiplexing themacross all available resources
Question 1: Granularity of Sharing
Frameworks can take turns accessing data on each node
Can resize frameworks shares to get
&utilizationinteractivity
Hadoop 1
Hadoop 1
Hadoop 1
Hadoop 1Hadoop 3
Hadoop 3 Hadoop 3
Hadoop 3
Hadoop 3
Hadoop 2
Hadoop 2Hadoop 2
Hadoop 2Hadoop 2
Hadoop 2
Hadoop 1
Hadoop 3
Hadoop 2Hadoop 3
Hadoop 1
Hadoop 2
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 13/86
Question 2: ResourceAllocation
Option: Global scheduler – Frameworks express needs in a specification
language, a global scheduler matchesresources to frameworks
• Requires encoding a framework’s semanticsusing the language, which is complex andcan lead to ambiguities
• Restricts frameworks if specification isunanticipated
Designing a general-purpose globalscheduler is hard
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 14/86
Question 2: ResourceAllocation
Nexus: Resource offers – Offer free resources to frameworks, let
frameworks pick which resources best
suit their needs+Keeps Nexus simple and allows us tosupport future jobs
- Distributed decisions might not beoptimal
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 15/86
Outline
• Nexus Architecture
• Resource Allocation
• Multi-Resource Fairness• Implementation
• Results
•
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 16/86
NEXUS ARCHITECTURE
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 17/86
Nexus slave
Nexus master
Hadoop v20 scheduler
Nexus slave
Hadoop job
Hadoop v20 executor
task
Nexus slave
Hadoop v19 executor
task
MPIscheduler
MPI job
MPIexecutor
task
Overview
Hadoop v19 scheduler
Hadoop job
Hadoop v19 executor
task
MPIexecutor
task
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 18/86
Nexus slaveNexus slave
Nexus master
MPI executor
task
Hadoopscheduler
Hadoop job
Resource Offers
MPIscheduler
MPI job
MPIexecutor
task
Pick framework to offer toResource
offer
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 19/86
Nexus slaveNexus slave
Nexus master
MPI executor
task
Hadoopscheduler
Hadoop job
Resource Offers
MPIscheduler
MPI job
MPIexecutor
task
Pick framework to offer toResource offer
ffer = { ,list of machine
}free_resources
:Example[ { , < ,node 1 2 CPUs 4
>},GB
{ , < ,node 2 2 CPUs 4>} ]GB
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 20/86
Nexus slave
Nexus master
Nexus slave
MPI executor
task
Hadoopscheduler
Hadoop job
Hadoopexecutor
Resource Offers
MPIscheduler
MPI job
MPIexecutor
task
Framework-specific schedulin
Pick framework to offer to
Launches & isolates execut
task
Resourceoffer
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 21/86
Resource Offer Details
• Min and max task sizes to controlfragmentation
• Filters let framework restrict offerssent to it
– By machine list
– By quantity of resources• Timeouts can be added to filters
• Frameworks can signal when to
destroy filters, or when they want
Using Offers for Data
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 22/86
Using Offers for DataLocality
We found that a simple policy calleddelay scheduling can give very high
locality: – Framework waits for offers on nodes
that have its data
– If waited longer than a certain delay,starts launching non-local tasks
–
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 23/86
Framework Isolation
• Isolation mechanism is pluggable dueto the inherentperfomance/isolation tradeoff
• Current implementation supportsSolaris projects and Linuxcontainers
– Both isolate CPU, memory andnetwork bandwidth
– Linux developers working on disk IOisolation
•
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 24/86
RESOURCE ALLOCATION
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 25/86
Allocation Policies
• Nexus picks framework to offerresources to, and hence controls howmany resources each framework can
get (but not which)• Allocation policies are pluggable to
suit organization needs, through
allocation modules
E l Hi hi l
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 26/86
Example: HierarchicalFairshare Policy
Facebook.com
Spam Ads
Job 3
Job 2
User 1
Job 1
User 2
Job 4
%100
CurrTime
%80%20
luster Share Policy
%20 %14%100
CurrTime
%6
CurrTime
%0
%70%30
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 27/86
Revocation
Killing tasks to make room for otherusersNot the normal case because fine-
grained tasks enable quick reallocationof resourcesSometimes necessary:
– Long running tasks neverrelinquishing resources
– Buggy job running forever
– Greedy user who decides to makes
his task long
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 28/86
Revocation Mechanism
Allocation policy defines a safe share foreach user
– Users will get at least safe share withinspecified timeRevoke only if a user is below its safe
share and is interested in offers
– Revoke tasks from users farthest abovetheir safe share
– Framework warned before its task iskilled
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 29/86
How Do We Run MPI?
Users always told their safe share
– Avoid revocation by staying below it
Giving each user a small safe sharemay not be enough if jobs need manymachinesCan run a traditional grid or HPC
scheduler as a user with a larger safeshare of the cluster, and have MPI jobsqueue up on it
– E.g. Torque gets 40% of cluster
xamp e: orque on
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 30/86
xamp e: orque onNexus
MPI Job
%40Safe share = 40%
MPI JobMPI Job
Torque
MPI Job
Facebook.com
Spam Ads
Job 1
Job 2
User 1
Job 1
User 2
Job 4
%40%20
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 31/86
MULTI-RESOURCEFAIRNESS
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 32/86
What is Fair?
• Goal: define a fair allocation of resources in the cluster betweenmultiple users
• Example: suppose we have: – 30 CPUs and 30 GB RAM
– Two users with equal shares
– User 1 needs <1 CPU, 1 GB RAM> pertask
– User 2 needs <1 CPU, 3 GB RAM> pertask
•
Definition 1: Asset
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 33/86
• Idea: give weights to resources (e.g. 1 CPU= 1 GB) and equalize value of resourcesgiven to each user
• Algorithm: when resources are free, offer towhoever has the least value
• Result:
– U1: 12 tasks: 12 CPUs, 12 GB ($24) – U2: 6 tasks: 6 CPUs, 18 GB ($24)
Definition 1: AssetFairness
PROBLEM
User 1 has < 50% of both CPUs and RAM
CPU
User1
User2%100
%50
%0RAM
essons rom
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 34/86
essons romDefinition 1
• “You shouldn’t do worse than if youran a smaller, private cluster equal in
size to your share”• Thus, given N users, each user shouldget ≥ 1/N of his dominating resource(i.e., the resource that he consumes
most of)•
D f 2 D i R
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 35/86
Def. 2: Dominant ResourceFairness
• Idea: give every user an equal share of herdominant resource (i.e., resource itconsumes most of)
• Algorithm: when resources are free, offer tothe user with the smallest dominant share( i.e., fractional share of the her dominantresource)
• Result: – U1: 15 tasks: 15 CPUs, 15 GB
– U2: 5 tasks: 5 CPUs, 15 GBCPU
User1
User2%100
%50
%0RAM
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 36/86
Fairness Properties→cheduler
↓ropertyAsset Dynamic CEEI DRF
Paretoefficiency
x x x x
-Single resourcefairness
x x x x
Bottleneckfairness
x x x
Share guarantee x x
Population
monotonicity
x x
-Envy freedom x x x
Resourcemonotonicity
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 37/86
IMPLEMENTATION
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 38/86
Implementation Stats
7000 lines of C++
APIs in C, C++, Java, Python, Ruby
Executor isolation using Linux
containers and Solaris projects
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 39/86
Frameworks
Ported frameworks:
– Hadoop (900 line patch)
– MPI (160 line wrapper scripts)
New frameworks:
– Spark, Scala framework for iterative jobs (1300 lines)
– Apache+haproxy, elastic web serverfarm (200 lines)
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 40/86
RESULTS
O h d
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 41/86
Overhead
Less than 4% seen in practice
Dynamic Resource
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 42/86
Dynamic ResourceSharing
Multiple Hadoops
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 43/86
Multiple HadoopsExperiment
Hadoop 1
Hadoop 2
Hadoop 3
Multiple Hadoops
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 44/86
Multiple HadoopsExperiment
Hadoop 1
Hadoop 1
Hadoop 1 Hadoop 1
Hadoop 1Hadoop 3
Hadoop 3 Hadoop 3
Hadoop 3
Hadoop 3
Hadoop 3
Hadoop 2
Hadoop 2Hadoop 2
Hadoop 2Hadoop 2
Hadoop 2
Hadoop 2 Hadoop 1
Hadoop 1
Hadoop 2
Hadoop 3 Hadoop 2
Hadoop 3
Results with 16
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 45/86
Results with 16Hadoops
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 46/86
WEB SERVER FARMFRAMEWORK
Web Framework
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 47/86
Load calculation
Nexus slave
Web FrameworkExperiment
Nexus master
Nexus slave
Web executor
task( )Apache
Scheduler (haproxy)Load gen framework
Load gen executor
task
httperf
Nexus slave
Web executor
task( )Apache
Load gen executor
task
HTTP requestHTTP request
Load gen
task task
executor Web executor
task( )Apache
HTTP request
resource offer
task
status update
W b F k R lt
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 48/86
Web Framework Results
F t W k
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 49/86
Future Work
Experiment with parallel programmingmodelsFurther explore low-latency services
on Nexus (web applications, etc)Shared services (e.g. BigTable, GFS)Deploy to users and open source
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 50/86
CLOUD COMPUTINGTESTBEDS
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 51/86
OPEN CIRRUS™: SEIZING THEOPEN SOURCE CLOUD STACK OPPORTUNITY A JOINT INITIATIVE SPONSORED BY HP, INTEL, AND YAHOO!: / / . /ttp o p e n cirru s o rg
Proprietary Cloud Computing
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 52/86
Applications
Application FrameworksMapReduce, Sawzall, Google AppEngine, Protocol Buffers
Hardware InfrastructureBorg
Software Infrastructure
VM Management
Job Scheduling
BorgStorage Management
GFS, BigTableMonitoring
Borg
Applications
Application FrameworksEMR – Hadoop
Hardware Infrastructure
Software Infrastructure
VM Management
EC2Job Scheduling
Storage Management
S3, EBSMonitoring
Borg
AMAZON
Applications
Application Frameworks.NET Services
Hardware InfrastructureFabric Controller
Software Infrastructure
VM Management
Fabric Controller Job Scheduling
Fabric Controller Storage Management
SQL Services, blobs, tables,queuesMonitoring
Fabric Controller
MICROSOFT
ublicly accessiblelayer
Proprietary Cloud Computingstacks
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 53/86
pen rrus ou ompu ng
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 54/86
pen rrus ou ompu ng Testbed
:ared research , applications , nfrastructure ( ),12K cores ata sets
obal services : , , .sign on monitoring store pen src stack ( , ,prs tashi hadoo ponsored by , , !P Intel and Yahoo (with additional support from NSF)
• , .9 sites currently target of around 20 in the next two years
O Ci G l
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 55/86
Open Cirrus Goals
• Goals• Foster new systems and services
research around cloud computing
• Catalyze open-source stack and APIs for
the cloud•
• How are we unique?• Support for systems research and
applications research
• Federation of heterogeneous datacenters
Open Cirrus Organi ation
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 56/86
Open Cirrus Organization
• Central Management Office, overseesOpen Cirrus
• Currently owned by HP
• Governance model• Research team• Technical team• New site additions• Support (legal (export, privacy), IT, etc.)
• Each site• Runs its own research and technical teams• Contributes individual technologies• Operates some of the global services
• E.g.• HP site supports portal and PRS• Intel site developing and supporting Tashi• Yahoo! contributes to Hadoop
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 57/86
Open Cirrus Sites
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 58/86
Open Cirrus Sites
Site Characteristics#Cores#Srvrs
PublicMemory Storage Spindles Network Focus
HP 1,024 256 178 3.3TB 632TB 1152 10G internal1Gb/s x-rack
Hadoop, Cells,PRS, scheduling
IDA 2,400 300 100 4.8TB 43TB+16TB SAN
600 1Gb/s Apps based onHadoop, Pig
Intel 1,364 198 145 1.8TB 610TB local60TB attach
746 1Gb/s Tashi, PRS, MPI,Hadoop
KIT 2,048 256 128 10TB 1PB 192 1Gb/s Apps with highthroughput
UIUC 1,024 128 64 2TB ~500TB 288 1Gb/s Datasets, cloudinfrastructure
CMU 1,024 128 64 2TB -- -- 1 Gb/s Storage, Tashi
Yahoo(M45)
3,200 480 400 2.4TB 1.2PB 1600 1Gb/s Hadoop ondemand
,2 074 ,746 ,029 .6 3 TB PBotal
Testbed Comparison
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 59/86
Testbeds
Open
Cirrus
IBM/GoogleTeraGrid PlanetLab EmuLab Open Cloud
Consortium
Amazon
EC2
LANL/NSF
cluster
Type of
researchSystems & services
Data-intensiveapplicationsresearch
Scientificapplications
Systemsandservices
Systems Interoperab.across cloudsusing openAPIs
Commer.use
Systems
Approach Federationof hetero-geneousdatacenters
A clustersupportedby Googleand IBM
Multi-siteheteroclusterssupercomp
A few 100nodeshosted byresearchinstit.
A single-sitecluster withflexiblecontrol
Multi-siteheterosclusters,focus onnetwork
Rawaccess tovirtualmachines
Re-use of LANL’sretiringclusters
Participants HP, Intel,IDA, KIT,
UIUC,Yahoo!CMU
IBM, Google,Stanford,
U.Wash,MIT
Manyschools
and orgs
Manyschools
and orgs
University of Utah
4 centers Amazon CMU, LANL,NSF
Distribution 7(9) sites1,746nodes12,074cores
1 site 11partnersin US
> 700nodesworld-wide
>300 nodesuniv@Utah
480 cores,distributed infour locations
1 site 1 site1000s of older, stilluseful nodes
Testbed Comparison
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 60/86
Open Cirrus Stack
+ +Compute network storage resources
+Power cooling
Management and
control subsystem
( )Physical Resource set Zoni service
: ( )Credit John Wilkes HP
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 61/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
,PRS clients each with their“ ”own physical data center
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 62/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
( . ., )Virtual clusters e g Tashi
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 63/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
BigData App
Hadoop
.1 Application running
.2 On Hadoop
.3 On Tashi virtual cluster.4 On a PRS
.5 On real hardware
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 64/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
BigData app
Hadoop
/Experiment
/save restore
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 65/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
BigData App
Hadoop
/Experiment
/save restore
Platformservices
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 66/86
Open Cirrus Stack
Zoni service
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
BigData App
Hadoop
/Experiment
/save restore
Platformservices
User services
Open Cirrus Stack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 67/86
Open Cirrus Stack
Zoni
Research Tashi NFS storageservice
HDFS storageservice
Virtual cluster Virtual cluster
BigData App
Hadoop
/Experiment
/save restore
Platformservices
User services
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 68/86
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 69/86
Open Cirrus Stack Tashi
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 70/86
Open Cirrus Stack - Tashi
• An open source Apache SoftwareFoundation project sponsored byIntel (with CMU, Yahoo, HP)
• Infrastructure for cloudcomputing on Big Data
• http://incubator.apache.org/projects/tashi
• Research focus:• Location-aware co-scheduling of
VMs, storage, and power.
• Seamless physical/virtualmigration.
• Joint with Greg Ganger (CMU),Mor Harchol-Balter (CMU), MilanMilenkovic (CTG)
T hi Hi h L l D i
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 71/86
C lu ste r
M a n ag e r
Tashi High-Level Design
Node
Node
Node
Node
Node
torage Service
irtualization Service
Node
Sche
duler
Cluster nodes are assumed to be commodity machines
Services are instantiated through virtual machines
Data location and power
information is exposed
to scheduler and services
CM maintains databases;and routes messages
decision logic is limited
Most decisions happen in
;the scheduler manages/ /compute storage power
in concert
The storage service aggregates the capacity of the commodity nodes
.to house Big Data repositories
Location Matters
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 72/86
Calculated (40 racks * 30 nodes * 2 disks)
0
50
100
150
200
250
300
Disk-1G SSD-1G Disk-10G SSD-10G
T h r o u g h p u t / d
i s k ( M B / s
Random Placement Location-Aware Placement
3 . 6
X
1 1 X
3 . 5
X
9 . 2
X
(calculated)
Open Cirrus Stack -
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 73/86
73
pHadoop
• An open-source Apache SoftwareFoundation project sponsored by
Yahoo!
• http://wiki.apache.org/hadoop/ProjectDesc
• Provides a parallel programming
model (MapReduce), a distributed filesystem, and a parallel database(HDFS)
projects are Open Cirrus
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 74/86
projects are Open Cirrussites looking for?
• Open Cirrus is seeking research in thefollowing areas (different centers will weightthese differently):
• Datacenter federation
• Datacenter management• Web services
• Data-intensive applications and systems
• The following kinds of projects are generallynot of interest:
• Traditional HPC application development
• Production applications that just need lots of cycles
• Closed source system development
How do users get access to
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 75/86
gOpen Cirrus sites?
• Project PIs apply to each site separately.
• Contact names, email addresses, and web linksfor applications to each site will be available
on the Open Cirrus Web site (which goes liveQ209) – http://opencirrus.org
–
• Each Open Cirrus site decides which users andprojects get access to its site.
• Developing a global sign on for all sites (Q2 09) – Users will be able to login to each Open Cirrus
site for which they are authorized using the
Summary and Lessons
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 76/86
Summary and Lessons
• Intel is collaborating with HP and Yahoo! toprovide a cloud computing testbed for theresearch community
• Using the cloud as an accelerator for interactivestreaming/big data apps is an importantusage model
• Primary goals are to• Foster new systems research around cloud
computing
• Catalyze open-source reference stack and APIs
for the cloud – Access model, Local and global services,Application frameworks
• Explore location-aware and power-awareworkload scheduling
• Develop integrated physical/virtual allocations tocombat cluster squatting
• Design cloud storage models
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 77/86
OTHER CLOUD COMPUTINGRESEARCH TOPICS:ISOLATION AND DC ENERGY
Heterogeneity in Virtualizedi
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 78/86
Environments
• VM technology isolates CPU and memory, butdisk and network are shared
– Full bandwidth when no contention
– Equal shares when there is contention
• 2.5x performance difference
EC2 smallinstances
Isolation Research
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 79/86
Isolation Research
• Need predictable variance over rawperformance
• Some resources that people have run
into problems with: – Power, disk space, disk I/O rate (drive,
bus), memory space (user/kernel),memory bus, cache at all levels (TLB,
etc), hyperthreading/etc, CPU rate,interrupts
– Network: NIC (Rx/Tx), Switch, cross-datacenter, cross-country
– OS resources: File descriptors, ports,
Datacenter Energy
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 80/86
Datacenter Energy
• EPA, 8/2007: – 1.5% of total U.S. energy consumption – Growing from 60 to 100 Billion kWh in 5
yrs
– 48% of typical IT budget spent onenergy
• 75 MW new DC deployments in PG&E’sservice area – that they know about!
(expect another 2x)• Microsoft: $500m new Chicago facility – Three substations with a capacity of
198MW
–200+ shipping containers w/ 2,000
servers each
Power/Cooling Issues
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 81/86
81
Power/Cooling Issues
First Milestone:DC E C ti
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 82/86
DC Energy Conservation
• DCs limited by power – For each dollar spent on servers, add
$0.48 (2005)/$0.71 (2010) for
power/cooling – $26B spent to power and cool servers
in 2005 grows to $45B in 2010
• Within DC racks, network equipmentoften the “hottest” components inthe hot spot
Thermal Image of TypicalCluster Rack
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 83/86
Cluster Rack
. . , . , . ,M K P a tte rso n A P ra tt P K u m a r
“ : - -From UPS to Silicon an end to end evaluation of” ,d ata ce n te r e fficie n cy In te lC o rp o ra tio n
RackSwitch
DC Networking andP
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 84/86
gPower
• Selectively power down ports/portions of netelements
• Enhanced power-awareness in the network stack – Power-aware routing and support for system
virtualization
• Support for datacenter “slice” power down andrestart
– Application and power-aware media access/control• Dynamic selection of full/half duplex• Directional asymmetry to save power,
e.g., 10Gb/s send, 100Mb/s receive
– Power-awareness in applications and protocols• Hard state (proxying), soft state (caching),protocol/data “streamlining” for power as well asb/w reduction
• Power implications for topology design – Tradeoffs in redundancy/high-availability vs. power
consumption – VLANs su ort for ower-aware s stem virtualization
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 85/86
8/6/2019 Cloud Computing 3999
http://slidepdf.com/reader/full/cloud-computing-3999 86/86
UC Berkeley
Thank you!
http://abovetheclouds.cs.berkeley.edu/