1 / 8

Some Context for This Session…

Some Context for This Session… . Performance historically a concern for virtualized applications By 2009, VMware (through vSphere ) and hardware vendors nearly eliminated the costs of storage, network, and memory virtualization

skah
Download Presentation

Some Context for This Session…

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Some Context for This Session… • Performance historically a concern for virtualized applications • By 2009, VMware (through vSphere) and hardware vendors nearly eliminated the costs of storage, network, and memory virtualization • With overheads near zero, new technologies in virtual deployments could sometimes beat physical counterparts • This session will focus on a diverse mix of extremelydemanding apps • Applications where we have proven performance with industry standard workloads and benchmarks • Not just speeds and feeds

  2. Virtual Machines ESX Server CPU Memory Storage Network Catalysts for Change - ESX 3.5 Platform Enhancements Scale Compatibility Performance • 64GB virtual RAM • Paravirtualization • Ubuntu • Windows Vista • 256 GB of physical RAM • Large memory pages • 10 GigE • Infiniband • TCP Segmentation Offload (TSO) • Jumbo Frames • NPIV Support • SATA devices

  3. 8-way vSMP and 255 GB of RAM per VM • 64 cores and 1 TB physical RAM • Wirespeed network access Catalysts for Change - ESX 4.0 Platform Enhancements VM Scale Up Virtual Machines • Virtual hardware scale out APP APP APP APP APP Hardware Scale Up OS OS OS OS OS ESX Hardware Assist Purpose Built Scheduler • Lowest CPU overhead CPU Hardware Assist Page Sharing Ballooning • Maximum memory efficiency Memory VMXNET3 VMDirectPath I/O Networking Storage stack optimization pvscsi • Less than 0.1 ms latency • Over 350,000 IOPS Storage Current NEW

  4. History Lesson on Application Performance 100% ESX 2 ESX 3 ESX 3.5 ESX 4.0 Overhead • 30% - 60% • 20% - 30% • <10% - 20% • <2% - 10% VM CPU • 1 vCPU • 2 vCPU • 4 vCPU • 8 vCPU Apps Supported VM Memory • 3.6 GB • 16 GB • 64 GB • 255 GB IO • <10,000 IOPS • 380 Mb • 800 Mb • 100,000 IOPS • 9Gb • >350,000 IOPS • 30 Gb ESX Version Source: VMware Capacity Planner analysis of > 700,000 servers in customer production environments

  5. Catalysts for Change - Hardware Improvements AMD 4M L2 cache 2010 AMD-V 10x faster AMD RVI released 2009 Intel VT-d 10x faster Intel EPT released 2008 AMD-V released Intel FlexPriority Released 2007 2006 Intel VT-d released Intel 4M L2 cache

  6. Catalysts for Change - Software Scalability Limited • VMware ESX Scaling: • Keeping up with core counts vSphere X Virtualization provides a means to exploit the hardware’s increasing parallelism vSphere 4 • Additional application scaling cost-prohibitive at some point

  7. Catalysts for Change: New Virtualization-based Architectures • vSphere-based deployments have options unavailable to physical servers: • On-loading multi-threaded IO drivers to efficiently use multiple cores • Scale out on a single host • Circumvents application scalability limitations • Improves memory locality of reference and increases cache efficiency • Hardware-accelerated network interrupt delivery • Strict memory encapsulation into NUMA nodes

  8. Catalysts for Change - Summary • VMware has made dramatic improvements in the performance of its virtualization platform • Hardware vendors have accelerated the efficiency of virtual workloads • vSphere provides flexibility that can allow administrators to circumvent application limitations • vSphere efficiently uses CPU in a way physical servers cannot • vSphere can meet and beat native application performance in many situations

More Related