Data Mobility – Caching Technologies

As part of his series looking at data mobility for the hybrid cloud, Chris Evans digs into the four main options for hybrid cloud caching. He reviews NAS, VM, database, and storage gateway caching, looking at the challenges involved for each.


Object Storage Critical Capabilities #3 – Searching, Indexing and Metadata

Chris Evans has been looking at object storage requirements in a series of posts. In this edition, he looks at how to handle object naming, the importance of user and system metadata, and the importance of search. For this last item, Chris highlights the Zenko multi-cloud object storage controller from Scality, which he saw at Storage Field Day in November.


Avere Systems is Acquired by Microsoft

A new year, a new acquisition by Microsoft. The Redmond software company acquired Avere Systems, with plans to integrate their hybrid cloud solutions into Azure. Chris Evans shares his thoughts on how this acquisition might effect the competitive landscape.


Performance Analysis of SAS/SATA and NVMe SSDs

Chris Evans reviews a performance analysis paper, breaking down the performance gains of NVMe vs SATA NAND disks in databases. This includes a look at real world applications. It shows the benefit of the reduced complexity and overhead of NVMe compared to SATA on otherwise identical storage media. As Chris notes, the lower CPU utlization and wait times have led to companies like E8 Storage and Excelero to develop new storage architectures. And in the HCI space, Scale Computing and X-IO are finding new ways to use the compute.


Has NVMe Killed off NVDIMM?

Chris Evans looks at the impact here of NVMe on the nascent NVDIMM market. Without needing to reorganize architectures and with OS support already in place, Chris sees NVMe has having a number of advantages. He still sees a place for NVDIMMs in the data center, but perhaps its place is much smaller than previously assumed.


Four Platforms When One Will Do?

After getting a review of their mid-range storage offerings at Storage Field Day last month, Chris Evans tries to makes sense of Dell EMC’s portfolio. This includes the SC-series, Unity, VNX and EqualLogic. The last two will be phased out as the result of the Dell EMC merger. Chris breaks down the rationale behind keeping two distinct mid-range lines, highlighting major differentiation like file protocols and deduplication support.


Is NetApp Becoming A Service Provider?

In this post, Chris Evans makes the argument that NetApp has fundamentally changed as a company, from a single product and platform to full on data management. As an example, he points to their Cloud Orchestrator service.


HPE Brings Nimble Skynet to 3PAR Arrays

Chris Evans considers how HPE will integrate Nimble Storage’s InfoSight into its 3PAR lineup after the former company’s acquisition. The challenge he sees is that Nimble Storage had a vast catalog of data points to pull from, having collected 4000 different metrics every five minutes from their arrays since 2010. If 3PAR doesn’t have that deep bench of data to pull from, it may take a while for InfoSight to derive valuable insights.


Accelerating Workloads with NetApp Plexistor

Chris Evans was surprised to learn that NetApp had acquired Plexistor, a company he had first seen at Storage Field Day in 2016. He thinks NetApp got a bargain in the acquisition, getting a storage class memory solution in their Plexistor SCM. Chris gives a quick overview how this could fit into NetApp’s portfolio.


VMAX – The Mainframe of Storage

In this piece, Chris Evans compares Dell EMC’s VMAX storage array to classic mainframe computing. This may sound like a backhanded compliment, but is actually astute praise. Like the mainframe, VMAX caters to legacy applications that need industry leading high availability and performance. In that niche, the new All-Flash VMAX array performs excellent, and Chris would recommend for the right legacy workloads.


Storage Field Day 14 Preview: E8 Storage

Chris Evans gives a preview of what he saw from E8 Storage at Storage Field Day last month. They use a disaggregated approach, combining the impressive parallelism of NVMe with RoCE to offer extremely low latency and high throughput on commodity storage hardware. Make sure to check out their full presentation from Storage Field Day to get a full overview.


Storage Field Day 14 Preview: Scality

Scality returns for their second Storage Field Day experience this week. Chris Evans will be there and wrote up a preview of the company ahead of the event. In the post, he provides a quick overview of what exactly is object storage, and how Scality differentiates within the category.


Storage Field Day 14 Preview: EMC

Chris Evans will be at Storage Field Day next month and he has a preview up looking at what to expect from Dell EMC. He takes a look at the company’s diverse portfolio of storage options, and hopes to get some clarity on how Dell EMC will better unify their midrange offerings at the event.


Can Violin Systems Successfully Rise from the Ashes of Violin Memory?

On his Storage Unpacked podcast, Chris Evans spoke to the CEO of Violin Systems, Ebrahim Abbasi. They discuss the future of the company coming out of their legacy with Violin Memory, and discussed focusing on tiering, scale-out and software-defined offerings for new product. Chris breaks down how he sees Violin Systems leveaging each of these in the competitive all-flash market.


It’s Time for Hard Drives to Join Tape In The Archive Tier

Mechanical hard drives are wonders of engineering. Despite proclamations of potential capacity limits, we’ve seen companies innovate to produce increasingly dense drives. But Chris Evans makes a compelling case here why spinning disks should be relegated to archiving going forward. This is because while throughput has increased much more slowly than capacity, we’ve actually seen latency get 50% worse in the last ten years. Combined with increased viability of cloud storage, Chris sees hard drives as viable for archive only.


Scale Computing Debuts HC3 in Google Cloud Platform

Scale Computing partnered with Google Cloud in their announcement to bring nested virtualization for their public cloud platform. Scale demonstrated this at Tech Field Day last week. Chris Evans looks at how with Scale’s HC3 Cloud Unity solution, the company is now able to offer Public Cloud as a DR location. Chris sees this fitting in with Scale Computing’s emphasis on simplicity offering organizations an easy path for low cost DR in a hybrid cloud.


Flash Diversity: High Capacity Drives from Nimbus and Micron

Chris Evans runs down some new high capacity drives from both Micron and Nimbus Data. Chris first got a look at some very different drives from Nimbus all the way back at the first Storage Field Day. The difference is rather stark, back then he saw some 800GB drives developed by Nimbus themselves. Today, Chris is looking at something more in the 50TB range. Micron’s 9200 series is no slouch either, topping off at 11TB in capacity and offering 270,000 random write IOPS.


Architecting IT | Scale Computing Moves Deeper Into the Enterprise With All-Flash HCI

Chris Evans looks back on Scale Computing, who he saw present at Storage Field Day in 2014, as they introduce their first all-flash system. Chris begins by stating that Scale Computing, “is one such company that has continued to demonstrate technical innovation and a continuing evolution in their platform,” and then goes on to discuss Scale’s all-flash system as well as their new software.


Pure1 META – Analytics for Pure Storage Arrays

After Pure Accelerate 2017, Chris Evans analyzes Pure Storage’s new storage array analytics solution, Pure1 META and its interesting testing applications.


Pure Accelerate: FlashArray Gets Synchronous Replication

Chris Evans writes about some important news coming out of Pure Accelerate. Pure Storage is now offering synchronous replication on its FlashBlade line. The feature is called ActiveCluster, and Chris got a technical deep dive on it at the Storage Field Day Exclusive event at Pure Accelerate. Chris does a great job running through the ActiveCluster architecture, and how those decisions will impact its adoption in the data center.