1° Sessione - Tecnologie hyperconvergenti e di virtualizzazione storage: VMware vSAN & HCI

Editor's Notes

• #4: ZZ – Start here
• #6: The momentum behind vSAN is continuing. As of Q3’16, we have now surpassed 5,500 customers and have seen over 150% bookings growth for 4 consecutive quarters.
• #7: At VMware we fundamentally believe HCI is a software-driven architecture – not a specific deployment model or hardware-locked configuration – that leverages two fundamental components: Hyper-Converged Software Industry-Standard Hardware The Hyper-Converged Software must deliver compute, storage, networking and management and this should be done from a tightly-integrated stack. The software-driven architecture allows the convergence of physical storage onto industry-standard x86 servers, enabling a building block approach with scale-out capabilities. The use of commodity x86 server and storage hardware allows datacenters to operate with agility on a highly scalable, cost-effective infrastructure.
• #8: So how is VMware approaching HCI? We are an industry-leading software company at our core. We are focused on providing the industry’s leading foundational components for HCI, converging compute, storage and management onto a single, integrated layer of software that can run on industry-standard hardware. VMware is the ONLY vendor who provides such level of integration. VMware Hyper-Converged Software is composed of three differentiating and best in class solutions: vSphere - the hypervisor that defines the industry standard vCenter Server – a unified and extensible management solution Virtual SAN – a radically simple, enterprise class storage solution that is uniquely embedded in the hypervisor. And today, we are proud to announce its 4th edition. This hyper-converged software is available through the broadest set of deployment options, from turnkey appliances to over one-hundred pre-certified servers through the Virtual SAN Ready Node program. We will discuss the deployment options and eco-system more in depth toward the end of this session.  
• #9: By adding additional capabilities like NSX for network virtualization and advanced management with vRealize, this becomes a complete SDDC solution.
• #10: So what makes VMware HCS differentiated among the various vendors in the HCI market? What makes our software stack the strongest in the industry? Customers are discovering that VMware is enabling HCI solutions that can deliver across 4 key areas:   It is Radically Simply – VMware is the only vendor that provides ONE, tightly integrated HCI software stack that can be run on your HW platform of choice, with no hardware lock-in. That choice – of over 100 pre-certified hardware stacks – mean there is no new hardware platform to learn. No new hardware vendor to get approved or new purchasing process to put in place. Buy and use the hardware you already know and most likely already use with vSphere today.   It offers the Highest Performance – VMware Hyper-Converged Software is optimized for all-flash and can deliver over one-hundred thousand IOPs per all-flash node with sub-millisecond latency. The hypervisor-embedded storage solution means we have an extremely short IO path and eliminate the numerous hops and delays associated with a virtual storage appliance.   It is Lowest Cost – We continue to deliver over 50% lower TCO compared to traditional storage. And thanks to the latest storage efficiency features in Virtual SAN, we can deliver all-flash storage for as low as $1 per usable GB. You no longer have to face the dilemma of picking either performance or cost.   It supports Any App, Any Scale – Designed for the future in mind, VMware hyper-converged software is capable of over 6M IOPS per cluster, and is designed to support any application. In addition to powering virtual desktops and other business critical applications today, Virtual SAN is being designed to run new cloud native applications and containers in the future.   We are thrilled to bring these amazing value propositions to our customers. So how are customers using Virtual SAN?
• #11: Virtual SAN provides both enterprise-class scale and performance as well as new capabilities that broaden the applicability of the now proven Virtual SAN technology to a wide variety of use cases at businesses of all types and sizes. With the benefits we just looked at, Virtual SAN is suited to be the storage for all your VMs. The most common use cases we hear from customers include business-critical applications, end user computing (VDI), disaster recovery, and remote office/branch office (ROBO). As Virtual SAN continues to get deployed for all types of VMs, we are seeing customer and market adoption grow as well…
• #12: Here you can see the public data points on our number of total customers plotted against the next largest HCI vendor in the market and the public customer counts they’ve announced. We estimate that we acquired over 70% more new customers in the second half o 2015 versus the other vendor. In addition, we shipped over 20,000 units based on CPU count in Q4’15 alone, outpacing any other HCI leading vendors. And finally, we saw nearly 200% year-over-year growth in Q4’15 thanks to strong customer demands across a broad set of verticals and use cases. It is plain and simple – Vmware IS the market leader in HCI. #1 in today customers and customer acquisition rate; #1 in numbers of units deployed; #1 in revenue growth. And, we have no plans to slow down. This is a perfect time to invite principle architect, and one of my favorite speakers, Rawlinson Rivera to cover what’s our brilliant engineering team has delivered in the latest version of Virtual SAN SOURCE: Publicly available customer counts from VMware press and Nutanix S-1 filing from Dec 2015. Quarterly data not available, so assumed linear growth for quarters in between provided data. Data is as follows, with bold data points being the publicly provided numbers and the rest being calculated assuming linear growth between the two nearest public points. These quarters are assuming a fiscal calendar year, but Nutanix quarters end a month later (FY ends July). The chart is being conservative and showing them hit their numbers a month earlier than they did.   Nutanix Virtual SAN Q4'11 0 0 Q1'12 35 0 Q2'12 70 0 Q3'12 106 0 Q4'12 141 0 Q1'13 176 0 Q2'13 211 0 Q3'13 354 0 Q4'13 497 0 Q1'14 639 0 Q2'14 782 400 Q3'14 923 800 Q4'14 1215 1200 Q1'15 1507 1600 Q2'15 1799 2000 Q3'15 2144 2500 Q4'15 2489 3000
• #13: A top priority for us to provide the broadest set of deployment options for VMware Hyper-Converged Software (VMware HCS), and we do that by maintaining the industry’s strongest ecosystem. As we look ahead, we have two clear HCI deployment options A focused-set of engineered HCI appliances delivering a turnkey experience An extremely broad set of pre-certified hardware platforms with optional software and licensing The decision point is simple and comes down to customers prioritizing deployment speed or deployment flexibility For customers that prefer a turnkey, deeply-integrated appliance, then stay tuned later next week for a exciting, game-changing announcement from the EMC Federation. For customers that want to build on the server infrastructure and experience they already have in place, then we are looking forward to offer even more options through our Virtual SAN Ready Node program
• #15: Virtual SAN is VMware’s software-defined storage solution, built from the ground up for vSphere virtual machines. It abstracts and aggregates locally attached disks in a vSphere cluster to create a storage solution that can be provisioned and managed from vCenter and the vSphere Web Client. Virtual SAN is hypervisor-converged – that is – storage and compute for VMs are delivered from the same x86 server platform running the hypervisor. It integrates with the entire VMware stack, including features like vMotion, HA, DRS etc. VM storage provisioning and day-to-day management of storage SLAs can be all be controlled through VM-level policies that can be set and modified on-the-fly. Virtual SAN delivers enterprise-class features, scale and performance, making it the ideal storage platform for VMs.
• #18: The rapid innovation around Virtual SAN has not slowed down. We are excited to be releasing our 4th edition of Virtual SAN, after introducing the first version only 24 months ago. With each update we’ve delivered greater value to our customers through key new features, really positioning us to be now ready to handle any virtual machine workload. With Virtual SAN 6.1 in September, we introduced key availability features like Stretched Cluster for 0-minute RPO and near-continuous protection and, for non-Stretched Cluster environments, we lowered the recovery point objective (RPO) with vSphere replication on Virtual SAN to only 5 minutes. In addition, key management features were added included the integrated Health Service solution with new monitoring capabilities. Virtual SAN 6.2 represents another major advance. In this release we are adding the much anticipating suite of space efficiency features, which includes deduplication, compression and erasure coding. In addition, new Quality of Service controls help guarantee SLAs and we continue to enhance the management capabilities.
• #19: Benefits: Eliminate cost and complexity of a separate physical array by using vSAN for physical/non-virtualized/multi-hypervisor workloads Expand Business Critical apps use case: supported Windows Server Failover Clustering (WSFC … the new MSCS) with shared storage Overview: Non-virtualized workloads and apps like SQL using Failover Clusters (WSFC) can utilize vSAN for storage (using iSCSI initiators) Export VMDKs as storage (LUN) for iSCSI targets managed through policies Core vSAN functionality available, including dedupe, compression, etc. Managed through vCenter
• #20: The 3 C’s this brings… 1. Cost. Allows customers with old slow switching (Sometimes 100Mbps!) to use 10Gbps for vMotion and vSAN using a pair of low cost Cat6 or TwinAX cables. 2. Complexity. No need to route the vSAN storage network over the WAN, and reduces the need for multi-cast configurations, and VLAN’s for vMotion and vSAN. 3. Compliance. Large enterprises have concerns with exposing storage networks even over encrypted VPN or private MPLS networks. By separating out the Witness traffic from the vSAN data traffic this reduces the footprint for remotely accessing data. Note this function can be used even without cross over.
• #21: Benefits Accelerate infrastructure responsiveness and SLAs with a fully scalable, automated, and simplified operational management Overview Full featured vSAN PowerCLI cmdlets to enable full automation Enterprise-class automation that brings cloud management flexibility to storage by supporting REST APIs Beyond our existing API, UI, CLI options PowerCLI support is being updated in this release. The following are planned to be supported in this release. vSAN Cluster Settings vSAN Disk and Disk Group Fault Domain and Stretched Cluster/ROBO Host Maintenance and Cluster Shutdown Health Check of vSAN Cluster Health Issue Remediation Resyncing Components Capacity Reporting Proactive Tests and Results
• #22: vSAN will continue integrating with the latest enhancements in hardware for drives, memory, controllers and servers. Also vSAN will take advantage of the latest hardware innovations in Next Gen NVRAM, NVDIMM, etc. 512e Drives will allow support 1.8TB 10K RPM drives for Hybrid vSAN. Storage industry is hitting capacity limit with 512N sector size used currently in rotating storage media. To address this issue, storage industry has proposed new Advanced Format drives which use 4K native sector size. These AF drives allows disk drive vendors to build high capacity drives and also provide better performance, efficient space utilization and above all, improved reliability and error correction capability. Large capacity SSDs mainly leverage 512n which remains fully supported. Refer to the vSAN HCL for the latest drive support.
• #24: Virtual SAN is managed through the vSphere Web Client and vCenter Server Virtual SAN is enabled at a cluster level and can be done by clicking a radio button Management is all done through the web client using the same toolset that is used for vSphere Not only does this not require a new management tool or training, but it enables anyone who knows vSphere to manage Virtual SAN
• #25: This is where it all starts, without Storage Policy Based Management many of the VMware HCS advantages would not be possible! While sometimes overlooked, Storage Policy Based Management is the key enabler for simple, software-defined storage! VMware’s Storage Policy Based Management framework allows you to define storage requirements on a per-VM basis, based on the needs of the applications running in the VMs. Simply define the amount of capacity, performance and availability for each VM, and then Virtual SAN matches those defined requirements to underlying storage infrastructure. Unlike traditional external storage, where provisioning is done at storage array layer, in a more rigid-hardware-centric way… Virtual SAN puts the application in charge. The end result, no more LUNs or Volumes … simplifying storage management and saving you time and money.
• #26: Virtual SAN integrates with the entire VMware stack, including features like vMotion, HA, DRS etc. Virtual SAN can leverage VMware vSphere Data Protection™, and vSphere Replication™ for data protection, backup, replication, and disaster-recovery (DR) purposes. Virtual SAN interoperates with vRealize Automation and can be deployed in conjunction with VMwareHorizon® View™ in VDI environments and vCenter Site Recovery Manager in DR environments.
• #28: Virtual SAN Ready node program is the only program to deliver a broad set of certified HW from all leading server vendors. This is a key driver behind the rapid customer adoption and our market leadership position. These hardware configurations form the foundation for a VMware HCS-powered HCI solution With this launch, the deployment choices are being expanded even further. Partners are now enabling customers to design VMware HCS-based HCI systems with maximum flexibility of hardware, software, licensing and support. Customers can choose to factory install the VMware hyper-converged software (HCS), namely vSphere and Virtual SAN. Customers can use their existing vSphere and VSAN licenses, or buy new licenses from the OEM vendor. All OEMs offer the flexibility of perpetual licenses that are node-transferrable, while some OEMs may also offer embedded licenses, which are fundamentally tied to the hardware system Customers can continue to purchase support from VMware, or leverage a single-vendor model by getting support for both hardware and software from their OEM of choice. These options are designed to simplify your life: Choose from the server vendor you already use … no new hardware to learn … no new procurement process to establish … no new vendor relationships to form Reduce deployment time by choosing from systems that are pre-installed (and in some cases partially pre-configured at the factory) with vSphere and Virtual SAN Receive the software licenses together or choose to use your own And finally, there are options for partners to provide one support number of hardware and software
• #33: All Flash Only. “High level description” Dedupe and compression happens during destaging from the caching tier to the capacity tier. You enable “space efficiency” on a cluster level and deduplication happens on a per disk group basis. Bigger disk groups will result in a higher deduplication ratio. After the blocks are deduplicated they will be compressed. A significant saving already, combined with deduplication and the results achieved can be up to 5x space reduction, of course fully dependent on the workload and type of VMs. “Lower level description” Compression (LZ4) would be performed during destaging from the caching tier to the capacity tier. 4KB is the block size for deduplication. For each unique 4k block compression would be performed and if the output block size is less than or equal to 2KB, a compressed block would be saved in place of the 4K block. If the output block size is greater than 2KB, the block would be written uncompressed and tracked as such. The reason is to avoid block alignment issues, as well as reduce the CPU hit for decompressing the data which is greater than compression for data with low compression ratios. All of this data reduction is after the write acknoledgement. 15% at most less than 5% greater resource usage. Deduplication domains are within each disk group. This avoids needing a global lookup table (significant resource overhead), and allows us to put those resources towards tracking a smaller and more meaningful block size. We purposefully avoid dedupe of “write hot data” In the cache, or decompressing uncompressible data significant CPU/memory resources can avoid being wasted. Note: Feature is supported with stretch clusters, ROBO edition
• #34: Sometimes RAID 5 and RAID 6 over the network is also referred as erasure coding. In this case RAID-5 requires 4 hosts at a minimum as it uses a 3+1 logic. With 4 hosts 1 can fail without data loss. This results in a significant reduction of required disk capacity. Normally a 20GB disk would require 40GB of disk capacity, but in the case of RAID-5 over the network the requirement is only ~27GB. There is another option if higher availability is desired Use case Information: Erasure codes offer “guaranteed capacity reduction unlike deduplication and compression. For customers who have “no thin provisioning policies” have data that is already compressed and deduplicated or have encrypted data this offers “known/fixed” capacity gains. This can be applied on a granular basis (Per VMDK) using the Storage Policy Based Management system. 30% Savings. Note: All Flash VSAN only. Note: Not supported with stretched clusters Note: this does not require the cluster size be a multiple of 4, just 4or more.
• #35: with RAID-6 two host failures can be tolerated, similar to FTT=2 using RAID-1. In the traditional scenario for a 20GB disk the required disk capacity would be 60GB, but with RAID-6 over the network this is just 30GB. Note that the parity is distributed across all hosts and there is no dedicated parity host or anything like that. Again, this is sometimes by others referred to as erasure coding. In this case a 4+2 configuration is used, which means that 6 hosts is the minimum to be able to use this configuration. Use case Information: Erasure codes offer “guaranteed capacity reduction unlike deduplication and compression. For customers who have “no thin provisioning policies” have data that is already compressed and deduplicated or have encrypted data this offers “known/fixed” capacity gains. This can be applied on a granular basis (Per VMDK) using the Storage Policy Based Management system. 50% savings Note: All Flash VSAN only Note: this does not require the cluster size be a multiple of six, just six or more. Not supported with stretched clusters
• #37: Comparison based on a Virtual SAN system where a customer needs ~25TB of usable capacity. The details of the configurations created are below. On Hybrid you’ll need: 50TB Raw ~10x1.2TB (on 4-node cluster) 2x400gb Caching SSD ~104K IOPS On All Flash w/ 2x Dedupe & EC: 19.2TB Raw ~6x800gb Capacity SSDs (on 4-node cluster) 1 x 400gb Cache SSD ~198K IOPS
• #38: Almost 6x more flash capacity on Virtual SAN 19.2TB All Flash Raw Capacity on VSAN (3.2TB cache) 16TB SATA HDDs with 3.8TB of Flash tier
• #41: Virtual SAN enables both all-flash and hybrid flash architectures. Irrespective of the architecture, there is a flash-based caching tier which can be configured out of flash devices like SSDs, PCIe cards, NVMe etc. The flash caching tier acts as the read cache/write buffer that dramatically improves the performance of storage operations. In All-Flash architecture, the flash-based caching tier is intelligently used as a write-buffer only, while another set of SSDs forms the persistence tier to store data. Since this architecture utilizes only flash devices, it delivers extremely high IOPs of up to 90K per host, with predictable low latencies. Reads will primarily come from the capacity tier, although if data was freshly written is “hot” and not de-staged it may read from the write cache tier. In the hybrid architecture, server-attached magnetic disks are pooled to create a distributed shared datastore, that persists the data. In this type of architecture, you can get up to 40K IOPs per server host.
• #43: StorageReview Article on NX-8150: http://www.storagereview.com/why_we_don_t_have_a_nutanix_nx8150_review
• #45: Virtual SAN allows you to either scale up or scale out your cluster Scale-out Add another server to add capacity and performance Each server contributes HDD to increase capacity and an SSD to contribute performance CLICK Scale-up Scale up for capacity by adding more HDDs Scale up for performance by adding more SSDs CLICK The end result is you are able to elastically and granular scale your storage system in a non-disruptive way Effective for building out what you need and modularizes your datacenter Eliminates need to make large upfront investments and gives you linear predictability for forecasting. Some of our customers really like this functionality given the trend towards distributed / scale-out applications T: This ability to flexibly scale and build out your storage system has become popular amongst our enterprise customers.
• #46: Stretched storage with Virtual SAN will allow you to split the Virtual SAN cluster across 2 sites, so that if a site fails, you would be able to seamlessly failover to the other site without any loss of data. Virtual SAN in a stretched storage deployment will accomplish this by synchronously mirror data across the 2 sites. The failover will be initiated by a witness VM that resides in a central place, accessible by both sites.
• #47: Virtual SAN 6.0 is the second-generation of VMware’s hypervisor-converged storage for VMs. Virtual SAN 6.0 capabilities are centered on delivering high performance and increased scale without compromising simplicity and cost-effectiveness. Most notably, Virtual SAN 6.0 now allows to create an all-flash architecture in which the VSAN datastore is carved entirely out of flash devices – i.e. SSDs are used for both caching and data persistence In VSAN 6.0, performance has been improved with 2x more IOPS on the hybrid configuration and 4x more IOPS on All-Flash, making VSAN ideal for high performance applications which require consistent response times and low latencies Virtual SAN 6.0 also features twice the scale, with the ability to scale to 64 hosts per VSAN cluster In addition, VSAN 6.0 can scale to run up to 200 VMs per host representing a 50% increase over the previous version It also features a new file system called the VSAN file system which provides efficient snapshots and clones Rack-awareness provides the ability to tolerate rack failures for a VSAN cluster that spans multiple racks Support for hardware controller-based checksums helps detects data corruption issues while support for hardware-based encryption can encrypt ‘data-at-rest’ for workloads that require the added level of security
• #50: KEY MESSAGES / TALK TRACK: These new features help simplify your tasks today but we are also looking ahead to make sure we create a platform that simplifies future IT demands To that end, we are excited to provide a glimpse into where we want to take Virtual SAN In the future we look forward to complementing the larger VMware Photon investment into Cloud Native Apps with Virtual SAN for VMware Photon We are designing what we think will be the best storage for DevOps What It Is Persistent storage for cloud native apps and next gen apps provided by VSAN with all existing features Tightly integrated with cluster managers (Mesos, Kubernetes, Swarm, etc) Managed solely via REST APIs and accessible to both IT admins and developers SPBM policies exposed as disk flavors in Photon platform Why It Matters End-to-end VMware: Tightly integrated solution for cloud native applications DevOps Focus: Promotes the developer to first class user in the datacenter and in control of storage provisioning activities (DevOps) Agile Storage Operations: Enables agility and scalability of storage operations with an API-friendly management model TRANSITION BACK TO YANBING
• #51: With VMware, you have three options for deploying and managing Containers and CNA. In all of them, vSAN is the default platform storage. The first two are brown-field use cases, i.e., in existing vSphere environment. The first is off-the-shelf vSphere – nothing else required. We already offer a Docker Volume Driver. You can use VMs as your Docker nodes. Docker, Docker Swarm, and other cluster managers all work out of the box (with a bit of API work we did recently). Ideal if you want to try containers, do some test&dev, etc. If you want a production environments for containers on vSphere, then VIC is the way to go. VM-like isolation for containers; new and traditional apps can co-exist on the same platform. Same management stack and co-existence of DevOps and traditional IT tools. Finally, Photon Platform is a platform to run native containers on a slim ESX hypervisor variant. Trusted hypervisor, new operational model, including for storage. A new version of vSAN will be available as part of the Photon platform … it will remove the vSAN dependency on vCenter so that it can be managed, monitored completely through APIs like Photon.
• #52: Infrastructure is only one part of the scalable management quest. The other is Application lifecycle management. How do you provision your apps, and their storage, how do you adapt to changing requirements, monitor the QoS per App and react to issues. Today, Policy-based automation has completely revolutionized how customers manage their applications. No need for manual LUN provisioning and management. The admin specifies in term of descriptive policies what they need for their apps, now how to do it. The platform, vSAN for example, automatically enforces them, monitors and performs automatic remediation when needed.
• #55: vSAN for ROBO – Aligns with the new vSphere for ROBO offering of 25 VMs Pack with up to one pack per site Can distribute VMs across sites, but no more than 25 VMs per site $500/VM with vSAN Standard features and $800/VM with vSAN ADV vSAN can support 2-nodes cluster beginning with 6.1 – will require witness VM (for every 2-node site)