Case Study: Increase SQL Server Performance 20x · Case Study: Increase SQL Server Performance 20x Sushant Rao ... • NetApp FAS2240-4 at primary and secondary (24TB each) • Scheduled

Copyright©2017DataCoreSo4wareCorp.–AllRightsReserved. DataCore So*ware The Data Infrastructure So*ware Company 1

Case Study: Increase SQL Server Performance 20x

Sushant Rao Senior Director of Product Marke2ng

Copyright©2017DataCoreSo4wareCorp.–AllRightsReserved. DataCore So*ware

•  LocatedinJacksonCountyonthesouthernOregonborder

•  ServesanesFmatedpopulaFonof200Kover2.8Ksquaremiles

•  Receivesover300Kcallsperyear•  40Dispatcherswith12

administraFvestaff•  Workswith30agenciescovering

Fire,LawenforcementandEmergencyMedicalagencies

2

Emergency CommunicaEons of Southern Oregon


•  Mission-criFcalapplicaFon:CalldispatchapplicaFonbuiltonMSSQLServer

•  Enter911calldetails•  Querydatabasestoaccesscallandcasehistories•  FindtheclosestunittosendforhelpusingGPS

•  ApplicaFonwasunabletokeepupwithdispatcher•  Thedispatcherwouldhavetowaitforqueriesorcommandstofinish

•  Rootcauseanalysisshowedbo\leneckwasstorage•  MulFpleFmesperday,latencieswere200ms+

3

Key Challenge: SQL Server Performance


Simplified View of the Dispatch ApplicaEon

4

Call Taker inputs call information to database (create to entry)

Dispatcher receives priors and premise

information

Call summary sent to Responders’ phone

Call details sent to Responders’ laptops

Copyright©2017DataCoreSo4wareCorp.–AllRightsReserved. DataCore So*ware 5

•  LawEnforcement–DispatchBenchmarks(PriorityE&1)

•  Processedwithin90seconds,90%oftheFme(CreaFontoenter)

•  Processedwithin120seconds,95%oftheFme(CreaFontoenter)

•  EmergencyMedical–DispatchBenchmarks(PriorityE&1)

•  Processedwithin90seconds,90%oftheFme(CreaFontoentry)

•  Dispatchedtotheappropriateagency(s)within30seconds,90%oftheFme(Entrytodispatch)

•  Fire–DispatchBenchmarks(Emergencyfireincidents)

•  Processedanddispatchedwithin90seconds,90%oftheFme(creaFontodispatch)

•  CreatetoEntry=60seconds•  EntrytoDispatch=30seconds

•  Processedanddispatchedwithin120seconds,95%oftheFme(creaFontodispatch).

•  CreatetoEntry=90seconds•  EntrytoDispatch=30seconds

ECSO Call & Dispatch Goals


ESCO IT Infrastructure (Previous)

Virtualized Infrastructure •  1 host each at primary

and secondary •  25 VMs: Standard

business applications, Active Directory and Exchange

10 most critical VMs

Storage •  NetApp FAS2240-4 at

primary and secondary (24TB each)

•  Scheduled synchronization job periodically depending upon the importance of the data

•  For critical data, job ran every 15 minutes

•  Most data was 12 hours old

Physical Servers •  SQL Server was NOT

virtualized due to inadequate storage performance

•  SQL Server hosts were clustered (DAS)


So4ware•  ESXIacceleratorso4ware

(PernixData,Infinio)•  Notenoughperformance;didn’t

addressavailability

•  VMwarevSANso4ware•  Expensive;incompaFblewith

exisFnghardware•  Zerto

•  Didn’taddressperformance

Hardware•  Newstoragearrays

•  NetApp•  PureStorage•  NimbleStorage

•  Decentperformancebutnotahugeincrease

AlternaEves Considered


•  UsedIometer•  Blocksizes:2K,4K,8K,32K,64K,128K,and256K•  OpFons:RandomvssequenFal/Readvswrite

•  28totalruns•  2K,Random,Read•  2K,Random,Write(andsoon)

•  Note:SameVMwasusedintesFng,buthardwarewasdifferent•  NetApphadseparateserversandstorage•  DataCorehyperconvergedhadcombinedserverandstorage

•  SameserversgeneraFngI/Oloadhadtohandleit

8

TesEng I/O Performance


0

5,000

10,000

15,000

20,000

25,000

30,000

Average Median

IOPS

NetApp DataCore

050100150200250300350400450500

Average Median

Throughput(MB/s)

NetApp DataCore

Comparison of I/O Performance

421%increase

1,374%increase

715%increase

429%increase


•  Numbers•  InProducFon:20ximprovementinlatency(onaverage)

•  Previously,peaklatencywas200msandhitmulFpleFmesperday•  Currentpeakislessthan5ms

•  InLab:4xto14xperformanceincrease•  Insomecases100XimprovementinIOPSandThroughput

•  Anecdotes•  Dispatchersnolongercomplainabouttheinputlagtheywereexperiencing•  AbletovMoFonvirtualmachinestosecondarysiteformaintenanceandusers

unawaretheyarenolongerrunningonprimarysite

10

Summary of Performance Improvement


ESCO IT Infrastructure (Current) Hosts •  SQL Server

instances were virtualized

•  Went from 5 total servers (2 virtual + 3 physical) down to 2 virtualized hosts

Hosts •  Went from 3 total

servers (1 virtual + 2 physical) down to 1 virtualized host

Storage •  DataCore

Hyperconverged Virtual SAN (1 per host) with direct-attached storage

•  Data is synchronously mirrored to DR site

•  911 calls, radio traffic, and critical SQL Server data are using CDP

•  NetApp storage became tier 3 (archive)


LowerCosts•  60%+lessinfrastructure

•  Reducedhostsfrom8to3andeliminatedexternalstorageforproducFondata•  SaveonadministraFonFmeandPCSE

•  Saved~$100KonSQLServerlicenses•  Changetopercorelicensingwasgoingtocost~$120Kwithphysicalhosts

•  Eachphysicalhostwassingleprocessorwith12cores(5serversx12cores)

•  ByvirtualizingSQLServer,abletocontrolnumberofcores•  1VMwith8vCPUs–Coresand2VMswith4vCPUs–Cores(totalof16vCPUs–Cores)

•  Be\erperformanceenabledconsolidaFonofinstances•  Wentfrom5physicalserversto3VMs

•  ~$25Kinlicensecostsa4erwards12

Cost Benefits


DataCore Hyperconverged Virtual SAN

13


•  HyperconvergedhasbeenusedprimaryforVDIandROBO•  VDIenvironmentstypicallyrequire200KIOPS

•  DatabasesandotherTier1workloadsrequiremuchmoreI/Operformanceandlowlatency

•  1MIOPSormoreareneeded•  Keymetricislatency

14

I/O Performance determines Use Cases for Hyperconverged

Sources: IDC


SPC Performance Benchmark: Database I/O Workload

Criteria SPCBenchmarkIndustryStandardIndependentlyVeriLied&AuditedPeerReviewedCoversdifferenttypes&generationsoftechnologyMapsto“realworld”performance(OLTPdatabases)Showscostforachievingperformancelevel

Enterprise Vendors run SPC Benchmarks


DataCore™ Hyperconverged DataCore™ Hyperconverged

DataCore: First Hyperconverged product to run SPC-1

WorldRecordResults!

PERFORMANCE

1.2millionSPC-1IOPSTM

HighestProducFvity

PRICE-PERFORMANCE

$0.10PerSPC-1IOPSTM

UnmatchedEconomics

RESPONSE TIME

0.22milliseconds

UltraFastApplicaFons

High-Availability/MulF-Node/FibreChannel

Source: Storage Performance Council SPC-1 Benchmark as of Nov 3, 20161


0.80 0.76

1.00 0.99

0.22

0

0.25

0.5

0.75

1

DataCore

$0.10

HPE 3PAR 8450 $0.23

NetApp EF560 $0.30

Dell SC4020 $0.37

EMC VNX8000

$0.41

Latency (Milliseconds)

Increasing Price per SPC-1 IOPS

4.8

17

DataCore Hyperconverged is Faster than AFAs


DataCoreRanBothSPC-1DatabaseWorkload+StorageLoadatSAMETime!

All-inclusivePrice

(HA2Nodes)

Hyperconverged SPC-1

$115K

1.2M IOPS

SAN(OnlyPar@al

Price)

ServersNotIncludedinPrice

Traditional SPC-1

$177K

435K IOPS >>

<<

Source: Storage Performance Council SPC-1 Benchmark as of Nov 3, 20162


DataCore Benefits

Improved performance by

3X or more

IMPROVE PERFORMANCE

Reduced storage-related downEme by

50% of more

REDUCE DOWNTIME

Reduced storage-related spending by

25% or more

INCREASE UTILIZATION

In the first year,

posiEve ROI

SAVE MONEY

19


30,000+ DEPLOYMENTS WORLDWIDE

10,000+ Customers 10th Gen Product

Companies in all Industries & Sizes

So*ware-defined Storage & Hyperconverged

Technology: Storage VirtualizaEon & Parallel I/O

Main Offices •  Australia •  France •  Germany •  Japan •  UK •  USA

PROVEN. GLOBALLY.

20


•  TryDataCoreHyperconvergedVirtualSANinyourenvironment•  SpeedupcriFcalapplicaFons&consolidateyourinfrastructure

•  Schedulea15-minutelivedemowithoneofourSoluFonArchitectsh\p://info.datacore.com/LiveDemo

•  DownloadFreeTrialofDataCore™HyperconvergedVirtualSANh\p://www.datacore.com/resources/free-hyperconverged-virtual-san

21

Next Steps

Copyright©2017DataCoreSo4wareCorp.–AllRightsReserved. DataCore So*ware The Data Infrastructure So*ware Company 22

Thank You!

[email protected] www.datacore.com


Be\erAvailability•  RPOgoalof0(nodataloss)

•  WantedsynchronousreplicaFontoDRsite

•  RTOgoalof30mins•  SFlllongFmeforemergency

communicaFonstobedown•  WantedvMoFontoDRsitewhen

needed

•  Abletotestfailoverandfailback

LowerCosts•  ConFnueusingexisFngNetApp

storage(onlyayearandhalfold)•  Needednewhostswith

addiFonalcomputeandmemory•  Wantedasinglepaneofglassfor

management

Other Challenges (aside from storage latency)


Be\erAvailability•  SynchronousreplicaFonbetweenhostssoRPOis0(nolossofdata)•  RTOisFmetorestartVMs

IncreasedFlexibility•  AddedmemorytoincreaseCache•  AddingNVMEtoaddstorageFer•  SQLServersarenowvirtualized

24

Availability & Flexibility Benefits


•  UPSatprimarysitewastobereplaced•  Wouldknockoutpowerfor6-10hourstoprimarysite•  AllVMswere“vMoFoned”tosecondarysitein10minutes

•  UserswereabletoconFnueaccessingapplicaFons•  NoperformancedegradaFon(completelytransparenttothem)

•  OnceUPSwasreplaced,allVMswere“vMoFoned”backtoprimarysite•  Startedatmidnight,backtoprimaryby9am

•  Storageisnolongerasinglepointoffailure,withnodatalossduetooutages

25

Real-World Availability SituaEon

Documents

Case Study: Increase SQL Server Performance 20x · Case Study: Increase SQL Server Performance 20x Sushant Rao ... • NetApp FAS2240-4 at primary and secondary (24TB each) • Scheduled