75: GreyBeards talk persistent memory IO with Andy Grimes, Principal Technologist, NetApp

This episode of the Greybeards on Storage podcast features a conversation with NetApp’s Principal Technologist, Andy Grimes. Andy presented to hosts Ray Lucchesi and Howard Marks at Tech Field Day Extra at NetApp Insight 2018, where he presented on MAX Data. This software-defined persistent memory solution comes out of the assets from NetApp’s Plexistor acquisition. They discuss some of the technical details, how it is designed to use emerging persistent memory, and use cases for the enterprise.

Screaming IOP performance with StarWind’s new NVMeoF software & Optane SSDs

It’s pretty impressive when a Storage Field Day presentation can get Ray Lucchesi thinking about building a super computer. But that’s just what happened after seeing the impressive performance from StarWind NVMeoF initiator on CentOS and Windows Server. Using just 2-core in a Hyper-V environment, StarWind was able to achieve bare metal storage performance on Optane, getting ~551K 4K random write IOP performance at the 0.6msec RT and 2.26 GB/sec level. Not surprising that Ray called StarWind an “amazing company.”

Huawei presents OceanStor architecture at SFD15

Ray Lucchesi goes in depth on Huawei’s new OceanStor Dorado storage system presented at Storage Field Day 15. This new system can scale up to 16 controllers, supporting all flash storage configurations, and supports inline compression and deduplication for data reduction. Overall, Ray was impressed by Huawei’s ability to reduce serialization bottlenecks and looks forward to hearing more from them in the future.

Infinidat’s ~90% (average read hit) solution

Ray Lucchesi reviews Infinidat’s presentation from Storage Field Day 16, impressed by their incredible 90% read hit solution. Ray says that the real challenge Infinidat faces is figuring out how to perform as well as an all flash array with hybrid flash disk storage, and it seems like they’ve accomplished that goal.

Infinidat announces new storage, replication & backup appliances + cloud storage

Ray Lucchesi outlines Infinidat’s new storage replication and backup appliances discussed at Storage Field Day 16. This includes InfiniSync, designed to be a “black box” to survive any disaster that could befall a data center, InfiniGuard, a solution to poor RTO results, as well as Infinidat’s new public cloud storage option. Ray discusses the purpose of each product as well as their positives and negatives looking forward.

Hitachi Vantara HCP, hits it out of the park

Ray Lucchesi had been aware of Hitachi Vantara’s Content Platform, aka on-premises object store, for a while. But it wasn’t until their Tech Field Day Exclusive event that he realized how successful it was. HCP currently has over two thousand customers and is the #1 on-premises, object storage solution in the world. There’s even more information available on HCP in our extensive video series from the event, be sure to check it out.

Western Digital at SFD15: ActiveScale object storage

Western Digital and Ray Lucchesi are both names synonymous with storage. We had them both attending Storage Field Day earlier this year. Ray was particularly interested in the companies ActiveScale object storage, a result of the company’s Amplidata acquisition in 2015. Ray runs through how ActiveScale could be well suited for data analytics.

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

Ray Lucchesi and Howard Marks talked to Jonathan Howard on the most recent GreyBeards on Storage podcast. Jonathan is the Director of Technical Alliances at Commvault, and they discuss the company’s move into the secondary storage market with Hyperscale, and how that combines with their data management solutions.

A tale of two storage companies – NetApp and Vantara (HDS-Insight Grp-Pentahoo)

Ray Lucchesi heard competing visions from storage companies in transitions recently. At Insight, NetApp presented themselves as the data service provider for IT and a willingness to embrace the cloud. The newly created Hitachi Vantara’s philosophy is based around the move to IoT. Ray lays out each company’s goals, it’s problem with customers, and who is ultimately right in with these divergent approaches.

Axellio, next gen, IO intensive server for RT analytics by X-IO Technologies

At Storage Field Day, the delegates saw a technical deep dive on X-IO Technologies new edge computing platform, Axellio. Ray Lucchesi runs down some the notable components of the platform. This includes support for up to 460TB of raw NVMe in a 2U appliance, the ability to add two off-load modules for parallel computing or machine learning, and over 12 Million IO/sec with at 35µsec latencies.

There’s a new cluster filesystem on the block, Elastifile

Elastifile debuted their new file system at Storage Field Day last month, and Ray Lucchesi wrote up his thoughts. It’s designed to support thousands of nodes, exabytes of capacity, and infinite numbers of files, in an effort to make a better cluster file system. It was in development since 2013, and offers some impressive features, including compression, deduplication, and cloud storage tiering. It only caches metadata, and maintains consistency thanks to key-value consensus based algorithm called Bizur. Ray’s not sure how it’ll perform in terms of marketshare going forward, but thinks it shows a lot of great backend engineering to offer a competitive file system right out of the gate.

Hardware vs. software innovation – round 4

Ray Lucchesi considers Dell EMC’s decision to kill their DSSD NVMe storage device, and frames it in the continuing debate over hardware vs software innovation. Ray thinks it’s further evidence that we are in a software innovation cycle. As further evidence, recent releases by both Excelero and E8 Storage. Both are using commodity hardware to achieve high level performance, over 4 million IO/sec with ~120 to ~230µsec response times. It’s an interesting discussion, and Ray gives both sides their due.

4.5M IO/[email protected]µsec 4KB Read on 100GBE with 24 NVMe cards

Excelero Storage launched at Storage Field Day last week. Ray Lucchesi was in the audience, and got to see some interesting performance numbers from their NVMesh, their software defined block storage for Linux. Ray is definitely enthusiastic about what he saw, with good reason. Excelero showed off getting 4.5 million 4K random reads and 2.5 million 4K random writes on $13,000 worth of hardware, all with 0% target CPU usage. Check out the rest of Ray’s piece for the details.

Dreaming of SCM but living with NVDIMMs…

Ray Lucchesi give a look into a rather unique and developing part of enterprise storage: NonVolatile RAM. He walks through the history of the format, and how it has evolved to its current iteration. He finds that the current state has fairly developed hardware, but the software to take advantage of it is still developing. In some ways, it parallels where flash storage was when it first emerged as an affordable option.

Docker presents at Cloud Field Day 1 (CFD1)

Docker presents at Cloud Field Day 1 (CFD1)

Scality’s Open Source S3 Driver

Scality’s Open Source S3 Driver

Hedvig storage system, Docker support & data protection that spans data centers

Hedvig storage system, Docker support & data protection that spans data centers

GreyBeards talk Copy Data Management with Ash Ashutosh, CEO Actifio

GreyBeards talk Copy Data Management with Ash Ashutosh, CEO Actifio

Exablox, bring your own disk storage

Exablox, bring your own disk storage

Pure Storage FlashBlade well positioned for next generation storage

Pure Storage FlashBlade well positioned for next generation storage