61: GreyBeards talk composable storage infrastructure with Taufik Ma, CEO, Attala Systems

In this episode,  we talk with Taufik Ma, CEO, Attala Systems (@AttalaSystems). Howard had met Taufik at last year’s FlashMemorySummit (FMS17) and was intrigued by their architecture which he thought was a harbinger of future trends in storage. The fact that Attala Systems was innovating with new, proprietary hardware made an interesting discussion, in its own right, from my perspective.

Taufik’s worked at startups and major hardware vendors in his past life and seems to have always been at the intersection of breakthrough solutions using hardware technology.

Attala Systems is based out of San Jose, CA.  Taufik has a class A team of executives, engineers and advisors making history again, this time in storage with JBoFs and NVMeoF.

Ray’s written about JBoF (just a bunch of flash) before (see  FaceBook moving to JBoF post). This is essentially a hardware box, filled with lots of flash storage and drive interfaces that directly connects to servers. Attala Systems storage is JBOF on steroids.

Composable Storage Infrastructure™

Essentially, their composable storage infrastructure JBOF connects with NVMeoF (NVMe over Fabric) using Ethernet to provide direct host access to  NVMe SSDs. They have implemented special purpose, proprietary hardware in the form of an FPGA, using this in a proprietary host network adapter (HNA) to support their NVMeoF storage.

Their HNA has a host side and a storage side version, both utilizing Attala Systems proprietary FPGA(s). With Attala HNAs they have implemented their own NVMeoF over UDP stack in hardware. It supports multi-path IO and highly available dual- or single-ported, NVMe SSDs in a storage shelf. They use standard RDMA capable Ethernet 25-50-100GbE (read Mellanox) switches to connect hosts to storage JBoFs.

They also support RDMA over Converged Ethernet (RoCE) NICS for additional host access. However I believe this requires host (NVMeoF) (their NVMeoY over UDP stack) software to connect to their storage.

From the host, Attala Systems storage on HNAs, looks like directly attached NVMe SSDs. Only they’re hot pluggable and physically located across an Ethernet network. In fact, Taufik mentioned that they already support VMware vSphere servers accessing Attala Systems composable storage infrastructure.

Okay on to the good stuff. Taufik said they measured their overhead and it was able to perform an IO with only an additional 5 µsec of overhead over native NVMe SSD latencies. Current NVMe SSDs operate with a response time of from 90 to 100 µsecs, and with Attala Systems Composable Storage Infrastructure, this means you should see 95 to 105 µsec response times over a JBoF(s) full of NVMe SSDs! Taufik said with Intel Optane SSD’s 10 µsec response times, they see response times at ~16 µsec (the extra µsec seems to be network switch delay)!!

Managing composable storage infrastructure

They also use a management “entity” (running on a server or as a VM),  that’s used to manage their JBoF storage and configure NVMe Namespaces (like a SCSI LUN/Volume).  Hosts use NVMe NameSpaces to access and split out the JBoF  NVMe storage space. That is, multiple Attala Systems Namespaces can be configured over a single NVMe SSD, each one corresponding to a single  (virtual to real) host NVMe SSD.

The management entity has a GUI but it just uses their RESTful APIs. They also support QoS on an IOPs or bandwidth limiting basis for Namespaces, to control manage noisy neighbors.

Attala systems architected their management system to support scale out storage. This means they could support many JBoFs in a rack and possibly multiple racks of JBoFs connected to swarms of servers. And nothing was said that would limit the number of Attala storage system JBoFs attached to a single server or under a single (dual for HA) management  entity. I thought the software may have a problem with this (e.g., 256 NVMe (NameSpaces) SSDs PCIe connected to the same server) but Taufik said this isn’t a problem for modern OS.

Taufik mentioned that with their RESTful APIs,  namespaces can be quickly created and torn down, on the fly. They envision their composable storage infrastructure to be a great complement to cloud compute and container execution environments.

For storage hardware, they use storage shelfs from OEM vendors. One recent configuration from Supermicro has hot-pluggable, dual ported, 32 NVMe slots in a 1U chasis, which at todays ~16TB capacities, is ~1/2PB of raw flash. Taufik mentioned 32TB NVMe SSDs are being worked on as we speak. Imagine that 1PB of flash NVMe SSD storage in 1U!!

The podcast runs ~47 minutes. Taufik took a while to get warmed up but once he got going, my jaw dropped away.  Listen to the podcast to learn more.

Taufik Ma, CEO Attala Systems

Tech-savvy business executive with track record of commercializing disruptive data center technologies.  After a short stint as an engineer at Intel after college, Taufik jumped to the business side where he led a team to define Intel’s crown jewels – CPUs & chipsets – during the ascendancy of the x86 server platform.

He honed his business skills as Co-GM of Intel’s Server System BU before leaving for a storage/networking startup.  The acquisition of this startup put him into the executive team of Emulex where as SVP of product management, he grew their networking business from scratch to deliver the industry’s first million units of 10Gb Ethernet product.

These accomplishments draw from his ability to engage and acquire customers at all stages of product maturity including partners when necessary.

58: GreyBeards talk HCI with Adam Carter, Chief Architect NetApp Solidfire #NetAppHCI

Sponsored by:In this episode we talk with Adam Carter (@yoadamcarter), Chief Architect, NetApp Solidfire & HCI (Hyper Converged Infrastructure) solutions. Howard talked with Adam at TFD16 and I have known Adam since before the acquisition. Adam did a tour de force session on HCI architectures at TFD16 and we would encourage you to view the video’s of his session.

This is the third time NetApp has been on our show (see our podcast with Lee Caswell and Dave Wright and our podcast with Andy Banta) but this is the first sponsored podcast from NetApp. Adam has been Chief Architect for Solidfire for as long as I have known him.

NetApp has FAS/AFF series storage, E-Series storage and SolidFire storage. Their HCI solution is based on their SolidFire storage system.

NetApp SolidFire HCI Appliance

 

NetApp’s HCI solution is built around a 2U 4-server configuration where 3 of the nodes are actual denser, new SolidFire storage nodes and the 4th node is a VMware ESXi host. That is they have a real, fully functional SolidFile AFA SAN storage system built directly into their HCI solution.

There’s probably a case to be made that this isn’t technically a HCI system from an industry perspective and looks more like a well architected, CI  (converged infrastructure) solution. However, they do support VMs running on their system, its all packaged together as one complete system, and they offer end-to-end (one throat to choke) support, over the complete system.

In addition, they spent a lot of effort improving SolidFire’s, already great VMware software integration to offer a full management stack that fully supports both the vSphere environment and the  embedded SolidFire AFA SAN storage system.

Using a full SolidFire storage system in their solution, NetApp  gave up on the low-end (<$30K-$50K) portion of the HCI market. But to supply the high IO performance, multi-tenancy, and QoS services of current SolidFire storage systems, they felt they had to embed a full SAN storage system.

With other HCI solutions, the storage activity must contend with other VMs and kernel processing on the server. And in these solutions, the storage system doesn’t control CPU/core/thread allocation and as such, can’t guarantee IO service levels that SolidFire is known for.

Also, by configuring their system with a real AFA SAN system, new additional ESXi servers can be added to the complex without needing to purchase additional storage software licenses. Further, customers can add bare metal servers to this environment and there’s still plenty of IO performance to go around. On the other hand, if a customer truly needs more storage performance/capacity, they can always add an additional, standalone SolidFire storage node to the cluster.

The podcast runs ~23 minutes. Adam was very easy to talk with and had deep technical knowledge of their new solution, industry HCI solutions and SolidFire storage.  It’s was a great pleasure for Howard and I to talk with him again. Listen to the podcast to learn more.

Adam Carter, Chief Architect, NetApp SolidFire

Adam Carter is the Chief Product Architect for SolidFire and HCI at NetApp. Adam is an expert in next generation data center infrastructure and storage virtualization.

Adam has led product management at LeftHand Networks, HP, VMware, SolidFire, and NetApp bringing revolutionary products to market. Adam pioneered the industry’s first Virtual Storage Appliance (VSA) product at LeftHand Networks and helped establish VMware’s VSA certification category.

Adam brings deep product knowledge and broad experience in the software defined data center ecosystem.

56: GreyBeards talk high performance file storage with Liran Zvibel, CEO & Co-Founder, WekaIO

This month we talk high performance, cluster file systems with Liran Zvibel (@liranzvibel), CEO and Co-Founder of WekaIO, a new software defined, scale-out file system. I first heard of WekaIO when it showed up on SPEC sfs2014 with a new SWBUILD benchmark submission. They had a 60 node EC2-AWS cluster running the benchmark and achieved, at the time, the highest SWBUILD number (500) of any solution.

At the moment, WekaIO are targeting HPC and Media&Entertainment verticals for their solution and it is sold on an annual capacity subscription basis.

By the way, a Wekabyte is 2**100 bytes of storage or ~ 1 trillion exabytes (2**60).

High performance file storage

The challenges with HPC file systems is that they need to handle a large number of files, large amounts of storage with high throughput access to all this data. Where WekaIO comes into the picture is that they do all that plus can support high file IOPS. That is, they can open, read or write a high number of relatively small files at an impressive speed, with low latency. These are becoming more popular with AI-machine learning and life sciences/genomic microscopy image processing.

Most file system developers will tell you that, they can supply high throughput  OR high file IOPS but doing both is a real challenge. WekaIO’s is able to do both while at the same time supporting billions of files per directory and trillions of files in a file system.

WekaIO has support for up to 64K cluster nodes and have tested up to 4000 cluster nodes. WekaIO announced last year an OEM agreement with HPE and are starting to build out bigger clusters.

Media & Entertainment file storage requirements are mostly just high throughput with large (media) file sizes. Here WekaIO has a more competition from other cluster file systems but their ability to support extra-large data repositories with great throughput is another advantage here.

WekaIO cluster file system

WekaIO is a software defined  storage solution. And whereas many HPC cluster file systems have metadata and storage nodes. WekaIO’s cluster nodes are combined meta-data and storage nodes. So as one scale’s capacity (by adding nodes), one not only scales large file throughput (via more IO parallelism) but also scales small file IOPS (via more metadata processing capabilities). There’s also some secret sauce to their metadata sharding (if that’s the right word) that allows WekaIO to support more metadata activity as the cluster grows.

One secret to WekaIO’s ability to support both high throughput and high file IOPS lies in  their performance load balancing across the cluster. Apparently, WekaIO can be configured to constantly monitoring all cluster nodes for performance and can balance all file IO activity (data transfers and metadata services) across the cluster, to insure that no one  node is over burdened with IO.

Liran says that performance load balancing was one reason they were so successful with their EC2 AWS SPEC sfs2014 SWBUILD benchmark. One problem with AWS EC2 nodes is a lot of unpredictability in node performance. When running EC2 instances, “noisy neighbors” impact node performance.  With WekaIO’s performance load balancing running on AWS EC2 node instances, they can  just redirect IO activity around slower nodes to faster nodes that can handle the work, in real time.

WekaIO performance load balancing is a configurable option. The other alternative is for WekaIO to “cryptographically” spread the workload across all the nodes in a cluster.

WekaIO uses a host driver for Posix access to the cluster. WekaIO’s frontend also natively supports (without host driver) NFSv3, SMB3.1, HDFS and AWS S3  protocols.

WekaIO also offers configurable file system data protection that can span 100s of failure domains (racks) supporting from 4 to 16 data stripes with 2 to 4 parity stripes. Liran said this was erasure code like but wouldn’t specifically state what they are doing differently.

They also support high performance storage and inactive storage with automated tiering of inactive data to object storage through policy management.

WekaIO creates a global name space across the cluster, which can be sub-divided into one to thousands  of file systems.

Snapshoting, cloning & moving work

WekaIO also has file system snapshots (readonly) and clones (read-write) using re-direct on write methodology. After the first snapshot/clone, subsequent snapshots/clones are only differential copies.

Another feature Howard and I thought was interesting was their DR as a Service like capability. This is, using an onprem WekaIO cluster to clone a file system/directory, tiering that to an S3 storage object. Then using that S3 storage object with an AWS EC2 WekaIO cluster to import the object(s) and re-constituting that file system/directory in the cloud. Once on AWS, work can occur in the cloud and the process can be reversed to move any updates back to the onprem cluster.

This way if you had work needing more compute than available onprem, you could move the data and workload to AWS, do the work there and then move the data back down to onprem again.

WekaIO’s RtOS, network stack, & NVMeoF

WekaIO runs under Linux as a user space application. WekaIO has implemented their own  Realtime O/S (RtOS) and high performance network stack that runs in user space.

With their own network stack they have also implemented NVMeoF support for (non-RDMA) Ethernet as well as InfiniBand networks. This is probably another reason they can have such low latency file IO operations.

The podcast runs ~42 minutes. Linar has been around  data storage systems for 20 years and as a result was very knowledgeable and interesting to talk with. Liran almost qualifies as a Greybeard, if not for the fact that he was clean shaven ;/. Listen to the podcast to learn more.

Linar Zvibel, CEO and Co-Founder, WekaIO

As Co-Founder and CEO, Mr. Liran Zvibel guides long term vision and strategy at WekaIO. Prior to creating the opportunity at WekaIO, he ran engineering at social startup and Fortune 100 organizations including Fusic, where he managed product definition, design and development for a portfolio of rich social media applications.

 

Liran also held principal architectural responsibilities for the hardware platform, clustering infrastructure and overall systems integration for XIV Storage System, acquired by IBM in 2007.

Mr. Zvibel holds a BSc.in Mathematics and Computer Science from Tel Aviv University.

54: GreyBeards talk scale-out secondary storage with Jonathan Howard, Dir. Tech. Alliances at Commvault

This month we talk scale-out secondary storage with Jonathan Howard,  Director of Technical Alliances at Commvault.  Both Howard and I attended Commvault GO2017 for Tech Field Day, this past month in Washington DC. We had an interesting overview of their Hyperscale secondary storage solution and Jonathan was the one answering most of our questions, so we thought he would make an good guest for our podcast.

Commvault has been providing data protection solutions for a long time, using anyone’s secondary storag, but recently they have released a software defined, scale-out secondary storage solution that runs their software with a clustered file system.

Hyperscale secondary storage

They call their solution, Hyperscale secondary storage and it’s available in both an hardware-software appliance as well as software only configuration on compatible off the shelf commercial hardware. Hyperscale uses the Red Hat Gluster cluster file system and together with the Commvault Data Platform provides a highly scaleable, secondary storage cluster that can meet anyone’s secondary storage needs while providing high availability and high throughput performance.

Commvault’s Hyperscale secondary storage system operates onprem in customer data centers. Hyperscale uses flash storage for system metadata but most secondary storage resides on local server disk.

Combined with Commvault Data Platform

With the sophistication of Commvault Data Platform one can have all the capabilities of a standalone Commvault environment with software defined storage. This allows just about any RTO/RPO needed by today’s enterprise and includes Live Sync secondary storage replication,  Onprem IntelliSnap for on storage snapshot management, Live Mount for instant recovery using secondary storage directly  to boot your VMs without having to wait for data recovery.  , and all the other recovery sophistication available from Commvault.

Hyperscale storage is capable of doing up to 5 Live Mount recoveries simultaneously per node without a problem but more are possible depending on performance requirements.

We also talked about Commvault’s cloud secondary storage solution which can make use of AWS S3 storage to hold backups.

Commvault’s organic growth

Most of the other data protection companies have came about through mergers, acquisitions or spinoffs. Commvault has continued along, enhancing their solution while bashing everything on an underlying centralized metadata database.  So their codebase was grown from the bottom up and supports pretty much any and all data protection requirements.

The podcast runs ~50 minutes. Jonathan was very knowledgeable about the technology and was great to talk with. Listen to the podcast to learn more.

Jonathan Howard, Director, Technical and Engineering Alliances, Commvault

Jonathan Howard is a Director, Technology & Engineering Alliances for Commvault. A 20-year veteran of the IT industry, Jonathan has worked at Commvault for the past 8 years in various field, product management, and now alliance facing roles.

In his present role with Alliances, Jonathan works with business and technology leaders to design and create numerous joint solutions that have empowered Commvault alliance partners to create and deliver their own new customer solutions.

51: GreyBeards talk hyper convergence with Lee Caswell, VP Product, Storage & Availability BU, VMware

Sponsored by:

VMware

In this episode we talk with Lee Caswell (@LeeCaswell), Vice President of Product, Storage and Availability Business Unit, VMware.  This is the second time Lee’s been on our show, the previous one back in April of last year when he was with his prior employer. Lee’s been at VMware for a little over a year now and has helped lead some significant changes in their HCI offering, vSAN.

VMware vSAN/HCI business

Many customers struggle to modernize their data centers with funding being the primary issue. This is very similar to what happened in the early 2000s as customers started virtualizing servers and consolidating storage. But today, there’s a new option, server based/software defined storage like VMware’s vSAN, which can be deployed for little expense and grown incrementally as needed. VMware’s vSAN customer base is currently growing by 150% CAGR, and VMware is adding over 100 new vSAN customers a week.

Many companies say they offer HCI, but few have adopted the software-only business model this entails. The transition from a hardware-software, appliance-based business model to a software-only business model is difficult and means a move from a high revenue-lower margin business to a lower revenue-higher margin business. VMware, from its very beginnings, has built a sustainable software-only business model that extends to vSAN today.

The software business model means that VMware can partner easily with a wide variety of server OEM partners to supply vSAN ReadyNodes that are pre-certified and jointly supported in the field. There are currently 14 server partners for vSAN ReadyNodes. In addition, VMware has co-designed the VxRail HCI Appliance with Dell EMC, which adds integrated life-cycle management as well as Dell EMC data protection software licenses.

As a result, customers can adopt vSAN as a build or a buy option for on-prem use and can also leverage vSAN in the cloud from a variety of cloud providers, including AWS very soon. It’s the software-only business model that sets the stage for this common data management across the hybrid cloud.

VMware vSAN software defined storage (SDS)

The advent of Intel Xeon processors and plentiful, relatively cheap SSD storage has made vSAN an easy storage solution for most virtualized data centers today. SSDs removed any performance concerns that customers had with hybrid HCI configurations. And with Intel’s latest Xeon Scalable processors, there’s more than enough power to handle both application compute and storage compute workloads.

From Lee’s perspective, there’s still a place for traditional SAN storage, but he sees it more for cold storage that is scaled independently from servers or for bare metal/non-virtualized storage environments. But for everyone else using virtualized data centers, they really need to give vSAN a look.

Storage vendors shifting sales

It used to be that major storage vendor sales teams would lead with hardware appliance storage solutions and then move to HCI when pushed. The problem was that a typical SAN storage sale takes 9 months to complete and then 3 years of limited additional sales.

To address this, some vendors have taken the approach where they lead with HCI and only move to legacy storage when it’s a better fit. With VMware vSAN, it’s a quicker sales cycle than legacy storage because HCI costs less up front and there’s no need to buy the final storage configuration with the first purchase. VMware vSAN HCI can grow as the customer applications needs dictate, generating additional incremental sales over time.

VMware vSAN in AWS

Recently, VMware has announced VMware Cloud in AWS.What this means is that you can have vSAN storage operating in an AWS cloud just like you would on-prem. In this case, workloads could migrate from cloud to on-prem and back again with almost no changes. How the data gets from on-prem to cloud is another question.

Also the pricing model for VMware Cloud in AWS moves to a consumption based model, where you pay for just what you use on a monthly basis. This way VMware Cloud in AWS and vSAN is billed monthly, consistent with other AWS offerings.

VMware vs. Microsoft on cloud

There’s a subtle difference in how Microsoft and VMware are adopting cloud. VMware came from an infrastructure platform and is now implementing their infrastructure on cloud. Microsoft started as a development platform and is taking their cloud development platform/stack and bringing it to on-prem.

It’s really two different philosophies in action. We now see VMware doing more for the development community with vSphere Integrated Containers (VIC), Docker Containers, Kubernetes, and Pivotal Cloud foundry. Meanwhile Microsoft is looking to implement the Azure stack for on-prem environments, and they are focusing more on infrastructure. In the end, enterprises will have terrific choices as the software defined data center frees up customers dollars and management time.

The podcast runs ~25 minutes. Lee is a very knowledgeable individual and although he doesn’t qualify as a Greybeard (just yet), he has been in and around the data center and flash storage environments throughout most of his career. From his diverse history, Lee has developed a very business like perspective on data center and storage technologies and it’s always a pleasure talking with him.  Listen to the podcast to learn more.

Lee Caswell, V.P. of Product, Storage & Availability Business Unit, VMware

Lee Caswell leads the VMware storage marketing team driving vSAN products, partnerships, and integrations. Lee joined VMware in 2016 and has extensive experience in executive leadership within the storage, flash and virtualization markets.

Prior to VMware, Lee was vice president of Marketing at NetApp and vice president of Solution Marketing at Fusion-IO (now SanDisk). Lee was a founding member of Pivot3, a company widely considered to be the founder of hyper-converged systems, where he served as the CEO and CMO. Earlier in his career, Lee held marketing leadership positions at Adaptec, and SEEQ Technology, a pioneer in non-volatile memory. He started his career at General Electric in Corporate Consulting.

Lee holds a bachelor of arts degree in economics from Carleton College and a master of business administration degree from Dartmouth College. Lee is a New York native and has lived in northern California for many years. He and his wife live in Palo Alto and have two children. In his spare time Lee enjoys cycling, playing guitar, and hiking the local hills.