HOUSE OF CARDS IAAS WITHOUT STORAGE PERFORMANCE TESTING - HOWARD MARKS, DEEP STORAGE LEN ROSENTHAL, LOAD DYNAMIX
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
House of Cards IaaS without storage performance testing Howard Marks, Deep Storage Len Rosenthal, Load DynamiX 1
Your NOT So Humble Speaker 25+ years of consulting and writing for trade press Columnist/blogger at NetworkComputing.com Chief Scientist DeepStorage, LLC Independent test lab and analysts @DeepStorageNet on Twitter Hmarks@DeepStorage.Net 2
Today’s Agenda What do we mean by IaaS? Why IaaS is for the corporate datacenter? The three types of IaaS storage How IaaS makes life difficult for storage ▸ The I/O blender and other hypervisor effects ▸ Users choose their storage level ▸ Noisy neighbors have a lease Which makes storage testing all the more critical 3
Infrastructure As A Service? Self service meets virtualization Users create VM from service catalog through web interface ▸ CPU, memory, storage capacity, storage class Can be delivered several ways ▸ Public (eg: AWS, Microsoft Azure) ▸ Private (eg: VMware vCAC, Openstack, Cloudstack) Still in your data center for security and privacy 4
IaaS Drivers Agility ▸ Workload deployment shrinks from days to minutes ▸ Workloads billed by the hour • Ideal for seasonal, temporary workloads • Encourages users to release unneeded resources • Can lead to higher utilization Allows IT to compete with public cloud 5
Storage = IaaS Magic Scaling the compute side of IaaS is science ▸ Just manage the vCPU to core ratio Storage for IaaS is still an art form ▸ Storage costs vary over 10X for same capacity • Therefore, a large fraction of total costs ▸ Storage guy metrics like IOPS alien to users ▸ Poor visibility into how applications affect storage ▸ Many projects stall on storage performance 6
IaaS Storage Types Non-persistent ▸ Block storage that’s released when VM shutdown Persistent block storage ▸ Eg: Amazon EBS, Openstack Cinder ▸ Dynamically provisioned, traditional storage Object Storage ▸ The cheap stuff eg: Amazon S3 ▸ Requires application support or gateway 7
Block Storage in Classes Defined in orchestration layer ▸ Media • SSD, hybrid, spinning disk, Etc. ▸ Data protection • RAID level, replication, snapshot schedule ▸ Hopefully storage QoS • Minimum IOPS, IOPS quota • Priority Some limited to gold, silver, bronze 8
The Cloud’s Dark Side Cloud providers have no visibility ▸ Into how storage will be used ▸ Into the application • So APM tools limited value ▸ Into future requirements The user/customer buys 200GB of Silver ▸ That may be a WORN archive or real time trading database ▸ They may make it 2TB overnight ▸ They may have needed bronze or plutonium 9
Virtualization is Hard on Storage Disk drives like sequential workloads When multiple VMs do sequential I/O they’re multiplexed together That throws I/O into a blender… all I/O is now random I/O! 5 VMs each needing 100 IOPS might need 700-1000 IOPS when virtualized 10
Enter the Noisy Neighbor Pre virtualization storage performance managed by spindles ▸ More IOPS, more 15K RPM drives ▸ Each app had dedicated spindles In IaaS multiple VMs, from different users, share a volume When marketing re-indexes their database at 2PM others may suffer 11
Can’t I Just Throw Flash At It? Flash based SSDs do provide lots of IOPS ▸ But even all flash arrays have their limits • and most of us can’t afford them Hybrid solutions are application sensitive ▸ How much flash? How to deploy? Storage performance validation with load testing can reveal: ▸ How much flash you need ▸ How your system will react to new workloads, noisy neighbors ▸ Performance limits 12
Storage Performance Validation A key to successful IaaS Len Rosenthal VP, Marketing July 24, 2014 13
Who is Load DynamiX? Who we are Leader in Storage Performance Validation Putting your storage technologies to the test Mission Provide insight into storage infrastructure behavior to assure performance & optimize cost for public and private clouds History Market entry March 2009; Privately-held Headquarters Santa Clara, CA Products Software: Load DynamiX Enterprise Hardware: LDX Series load generation appliances 14
Storage Performance Validation Go Daddy Case Study Go Daddy Overview World’s largest web hosting provider: 5+ million accounts Over 50 million domain names under management “Load DynamiX 3300 employees; 600+ IT developers, 5 storage admins is like my smart phone – I can’t IT Challenge imagine my life Determine most cost-effective storage systems to support without it!” 31PB growing 30% annually with 99.999% uptime across 5 data centers Julia Palmer Manager, Solution Performance Multiple Load DynamiX 10G Series appliances Engineering Go Daddy Benefits Go Daddy Projected storage cost savings of 65% over 2 years Full confidence that their storage systems can support projected workloads Freedom to rapidly innovate with new technologies 15
Attaining IaaS Enlightenment Validate with Validate with freeware tools Test in (Iometer, etc.) production and pray If it ain’t broke don’t fix it! Evolution of Storage Performance Validation Evolution 16
Storage Performance Validation Process In 3 Simple Steps Iterate on different products & configurations 1 Create 2 Run 3 Analyze Model Tests Results Create storage Connect storage Find performance workload emulations systems to Load limits. Compare from production data DynamiX load latency, throughput & or use our sample generation appliance IOPS of different workloads. and run the tests. products and configs. Iterate on different model assumptions, scale Deploy In Production with Confidence 17
Go Daddy Production Workload Virtually identical to the one created in Load DynamiX 18
Storage Cost Optimization Driving cost optimization through technology validation Existing Vendor Baseline Vendor B 23% drop Add Compression OS & Disk Change 35% drop 56% drop Dedupe/Compression Cost per GB Vendor C 62% drop 56% drop Whitebox 67% drop Every transition validated with Load DynamiX Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 19
Benefits to Go Daddy by Using Load DynamiX Superior Storage ROI Pick most cost effective solution for workload Faster Validation Weeks to test in lab, not 12 months in Performance production Testing Confidence in user experience Freedom to Innovate Insight for Availability better decision- Storage Confidence in 5- making Engineer 9’s under load 20
Q&A Session Howard Marks Len Rosenthal DeepStorage.net Load DynamiX.com 21
You can also read