Virtual Machine Workloads: The Case For New Benchmarks For
Virtual Machine Workloads:The Case for New Benchmarks for NASvirtualT2M Vasily Tarasov, Dean Hildebrand, Geoff Kuenning, Erez Zadok Stony Brook University, IBM Research – Almaden, Harvey Mudd College
OverviewStorage CompaniesNetworkAttachedStorage!FastStorage Inc.ClientsNeverFail Corp.TroubleFree LLCBenchmark!2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS2
OverviewStorage CompaniesNetworkAttachedStorage!ClientsSlowStorage Inc.BenchmarkEpicFail Corp.RealityRestLess LLCBenchmark!2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS3
OverviewStorage CompaniesNetworkAttachedStorage!ClientsFastStorage Inc.SlowStorageBenchmarkNeverFailEpicFail Corp.RealityRestLess LLCTroubleFreeLLCBenchmark!But not the benchmarks!2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS4
OverviewStorage CompaniesNetworkAttachedStorage!ClientsFastStorage Inc.SlowStorageBenchmarkCreate benchmarks for NAS thataccurately generate workloadsNeverFailEpicFail Corp.originating in Virtual EnvironmentsRealityRestLess LLCTroubleFreeLLCBenchmark!But not the benchmarks!2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS5
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS6
Data Access Options for VMsFile system interface in VMs12Emulated block device Guest OS on-disk FSDistributed File Systemclient in guest OSVMWare Workstationwith VDI on Ext3VM-NASabcdVDI on VDI on VDI onNASSAN(VDI – Virtual Disk Image) DASVMWare ESXiwith VDI on VMFS2/15/2013VMWare ESXiwith VDI on a filerVirtual Machine Workloads: The Case for New Benchmarks for NAS7Pass-throughto DAS/SANVMWare ESXiwith SAN
Trends Shipped storage capacityNetwork-basedStorage Direct-attachedstorage10%Growth NAS – 60% SAN – 23%90%x86 server virtualizationVirtualizedservers[IDC 2010] (by 2014)[Gartner 2010] (by 2014)30%70%2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS8Non-virtualizedservers
Trends Shipped storage 10% Growth NAS – 60% SAN – 23%Virtualization90%VMsonNAS x86 server virtualizationVirtualizedservers[IDC 2010] (by 2014)NetworkAttachedStorage[Gartner 2010] (by 2014)30%70%2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS9Non-virtualizedservers
VM-NAS I/O StackReordering,Merging,Splitting, ApplicationsGuest OSVirtual File SystemMapping, Caching,Coalescing, .On-Disk File SystemBlock LayerController DriverHypervisor & VMController EmulatorNFS ClientRead/Writesizes,Attributecaching, VariousProposedOptimizationsNetworkNFS ServerHost OSVirtual File SystemOn-Disk File SystemBlock LayerController Driver2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS10
VM-NAS I/O StackReordering,Merging,Splitting, ApplicationsGuest OSVirtual File SystemMapping, Caching,Coalescing, .On-Disk File SystemBlock LayerController DriverHypervisor & VMController EmulatorNFS ClientRead/Writesizes,Attributecaching, VariousProposedOptimizationsNetworkNFS ServerHost OS1. Deep I/O stackVirtual File System2. Requests changeOn-Disk File System significantlyBlock LayerController Driver2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS11
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS12
Physical vs. tualMachineVirtualMachinePhysical MachineNFS /SMBNFS /SMBW1W2NASApplianceNASApplianceGPFS, WAFL, ZFSGPFS, WAFL, ZFSCurrent NASBenchmarks2/15/2013W1 New NASBenchmarksW2Virtual Machine Workloads: The Case for New Benchmarks for NAS13
Physical vs. tualMachineVirtualMachinePhysical MachineNFS /SMBNFS /SMBW1NASApplianceMeta-data opsSPECsfs2008: 72%Virtual setup: 1%2/15/2013NASApplianceGPFS, WAFL, ZFSGPFS, WAFL, ZFSCurrent NASBenchmarksW2W1 New NASBenchmarksW2Virtual Machine Workloads: The Case for New Benchmarks for NAS14
How to Evaluate VM-NAS? Run old benchmarks in VMs Cumbersome Hypervisors, VMs , OSes, applications Inflexible Many combinations of different workloads Expensive Equipment to run hypervisors and VMs 2/15/2013Create new benchmarksVirtual Machine Workloads: The Case for New Benchmarks for NAS15
Configuration Filebench: File, Web, Database serverApplication Jetstress (Mail MachineVirtualMachine OperatingSystem: Linux/Win File system: ext3/NTFS I/O scheduler: CFQ VM parameters: defaultPhysicalMachine Hypervisor: VMWare ESXi 5 NFS client: default settingsNAS ApplianceBack-end File System2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS16 Black box Linux GPFS tracingfacilities
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS17
ObservationsWorkload PropertyPhysical NAS clientsVirtual NAS clientsFile and directory countMany files and directoriesFew files per VMDirectory tree depthDeep and non-uniformShallow and uniformFile sizeLean towards small filesMulti-gigabyte, but sparseMeta-data operationsManyAlmost noneI/O synchronizationAsync and syncAll writes are syncIn-file randomnessWorkload-dependentIncreased randomnessCross-file randomnessWorkload-dependentPredictableI/O sizesWorkload-dependentIncreased and decreasedRead-modify-writeInfrequentMore frequentThink timeWorkload-dependentIncreased2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS18
Workload Features Read/write ratio I/O Size distribution Jump distance distribution LBA distance between two consecutive requestsOffset 013Web-serverDatabase-serverVirtual Machine Workloads: The Case for New Benchmarks for NAS19Mail-server(Win)
I/O Size, Jump Distance, Offset Popularity504030Percent20100Web ServerReadsWrites048 12 16 20 24 28 32 36 40 44 48 52 56 60 64I/O Size (KB)Percent100806040200Database Server-16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6Jump Distance 0138 10 12 14 1623456 7 8 9 10 11 12 13 14 15 15RNumber of accessesVirtual Machine Workloads: The Case for New Benchmarks for NAS20
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS21
T2M atures2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NASWorkloaddescription inBenchmark’slanguage22
T2M BenchmarksgenerateFeaturesstationaryworkloads – almost no chunking2. Block level tracesNFS tracesBut block-on-file.2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NASWorkloaddescription inBenchmark’slanguage23
Multi-VM Declarations Define hypervisors and VM:define hypervisor type esx51 {define vm type fileserver,instances 5 { }define vm type webserver,instances 2 { }}2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS24
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS25
Accuracy Metrics21 Non-virtualizedBenchmarkin VMVirtualizedBenchmarkCompareMonitored NAS parameters: 2/15/2013Reads/SecWrites/SecLatencyI/O UtilizationQueue lengthRequest size CPU UtilizationMemory usageInterruptsContext switchesWaiting processesVirtual Machine Workloads: The Case for New Benchmarks for NAS26
Single VM Accuracy (Web-Server)109876Error (%) 5432102/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NASRMSMaximum27
Multi-VM Accuracy14Reads/SecWrites/Sec12LatencyI/O Utillization10Queue Length8Request SizeMax Error (%)CPU Utilization6Memory Usage4InterruptsContext switchesh2Waiting processes012/15/20132345Number of VMsVirtual Machine Workloads: The Case for New Benchmarks for NAS67288
Outline1.2.3.4.5.6.2/15/2013BackgroundExperimental SetupWorkload CharacterizationBenchmark CreationBenchmark AccuracyConclusions and Future WorkVirtual Machine Workloads: The Case for New Benchmarks for NAS29
Conclusions Virtualization causes NAS workloads to change Current benchmarks are no longer representative Need for VM-oriented benchmarks Trace-based analysis of VM-NAS workloads Trace2Model for NFS traces Created a set of VM-benchmarks Accuracy within 10% on average2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS30
Future Work Explore other configurations The impact of VM-NAS parameters VM workload classification VM-specific workloads snapshotting, boot and update stormsEmulate I/O request transformations Multi-level trace analysis Higher flexibility2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS31
Virtual Machine Workloads:The Case for New Benchmarks for NASQ&AGoogle it: t2mpublic Vasily Tarasov, Dean Hildebrand, Geoff Kuenning, Erez Zadok Stony Brook University, IBM Research – Almaden, Harvey Mudd College2/15/2013Virtual Machine Workloads: The Case for New Benchmarks for NAS32
Network Attached Storage VMs on NAS . 2/15/2013 Virtual Machine Workloads: The Case for New Benchmarks for NAS 10 VM-NAS I/O Stack . GPFS, WAFL, ZFS GPFS, WAFL, ZFS NAS Appliance Application NFS /SMB Current NAS Benchmarks New NAS Benchmarks Physical Machine
May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)
Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .
On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.
̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions
Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have
Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được
series b, 580c. case farm tractor manuals - tractor repair, service and case 530 ck backhoe & loader only case 530 ck, case 530 forklift attachment only, const king case 531 ag case 535 ag case 540 case 540 ag case 540, 540c ag case 540c ag case 541 case 541 ag case 541c ag case 545 ag case 570 case 570 ag case 570 agas, case
Le genou de Lucy. Odile Jacob. 1999. Coppens Y. Pré-textes. L’homme préhistorique en morceaux. Eds Odile Jacob. 2011. Costentin J., Delaveau P. Café, thé, chocolat, les bons effets sur le cerveau et pour le corps. Editions Odile Jacob. 2010. Crawford M., Marsh D. The driving force : food in human evolution and the future.
