Itzikr's Blog

In the first post, we covered the richness of the PowerScale platform and why, it’s the leading scale-out, storage platform. It’s is now time to cover what we have just announced, but let’s start with the challenges we are facing today re Data. There are a few questions to ask when assessing whether your organization is equipped to manage the coming deluge of data:

• How much data do I have & where is it? Am I able to find it and bring it all under my control?
• Is my data secure? Do I have vulnerable shadow data out there?
• Is it stored in the right place & accessible? Do the right people have access at the right time? Are my data management tools self-service? Do I want non-IT personnel to manage data?
• Can my data be moved? Should it be moved? Who gets to move it?
• Is time wasted on manual tasks? Can I automate processes to free up time to focus on strategic initiatives vs. managing storage?

If the answer to any of the above is no, we have a growing ecosystem of data management software that can help you answer yes to all these questions and much more.

McKinsey conducted “The state of AI in 2020” study recently, and it turns out that AI is not just a buzz word anymore. They found “80% of people said that AI has helped increase revenue.” However, there is a problem. One of the biggest challenges to AI & Machine Learning adoption is scaling up. Traditional storage technologies are failing to keep up.

In addition, AI and GPU powered workloads demand speed and scale. Today, they are fragmented in data silos across the edge, the core, and the cloud. And because of this fragmentation, there is inconsistent performance, vendor lock-

in, unpredictable economics, and often low availability problems.

Based on the latest research from ESG we’re seeing that customers are building their data lakes on flash storage to overcome these challenges. In fact, in a recent survey, ESG found that more than 80% of the customers with production AI/ML environments tend to have most of their storage footprint on flash

storage.

To address this growth in the velocity, volume, and variety data, you need the enormous power and capacity of all-flash storage to drive demanding workloads for today and tomorrow. Today, we are introducing the new PowerScale F900 – the most powerful platform in the PowerScale family.

PowerScale All-Flash Nodes Comparison

It delivers on 3 pillars: More power and scale; more flexibility and choice; and support of a world-class eco-system of software and solutions For Power and Scale, we have an All-NVMe powered node which has the capability to drive demanding GPU-powered analytics and HPC applications. The F900 delivers up to 93 PB of raw capacity per cluster. For Flexibility and Choice, F900 can be seamless deployed in new clusters or added to existing PowerScale or Isilon clusters.

A recent Forrester Total Economic Indicator (TEI) study showed that the F900 powered by OneFS can deliver an ROI of up to 374% and a payback period of less than 6 months. Plus, it can be consumed either as an appliance or as an APEX Data Storage Services. Ultimately, it’s about the software, applications and workloads that power the business. The OneFS-powered PowerScale nodes can support a wide range of

applications that are tested and certified across a variety of workloads across all industries.

OneFS is complemented by a vast catalog of best-in-class solutions for backup, analytics, data management, migration services, cloud services, client applications and more.

It’s not just about speeds and feeds: Performance boosts are not just from the new hardware. We’ve optimized the 9th generation of OneFS to deliver more performance. OneFS 9.2 speeds up the F600 by up to 70% for sequential reads by taking advantage of multi-processor capabilities.

Ultimately, it’s about the power of OneFS that enables administrators to focus on managing their data, not the storage.

With PowerScale’s OneFS 9.0 and F200 and F600 Nodes… we add to our existing deduplication capabilities our first PowerEdge-based platforms providing inline data reduction, enabling our customers to provide capacity more economically for performance applications, at scale. This new capability complements our post-process deduplication engine for large archive data sets where policies can be set at the filesystem, or at the file level, depending on customer and application need. Customers also get more value through our inline data reduction and smart dedupe features available on the F900, F600, and F200. Customers can benefit from these data reduction features especially for

active-archive workloads such as AI/ML.

The new high-end All-NVMe F900 node offers more power & speed, delivering 30% faster sequential reads, support for NVIDIA GPUDirect, and multi-CPU server platform optimized with high density. It is ideal for high performance data access and next-gen workloads requiring high single stream throughput that can leverage NVMe flash. And it offers seamless connectivity through seamless interoperability with existing Isilon clusters.

With the latest OneFS 9.2, we have also powered up the F600 and F200, launched last year. There’s higher performance with up to 70% increase in sequential reads for F600 and up to 25% for sequential reads for the F200. Customers also get more flexibility through new drive options, and the ability to add these nodes to existing Isilon clusters non-disruptively. Finally, customers get data-at-rest encryption through self-encrypting drives (SED) on F200.

With addition of the NFS over RDMA capability, OneFS 9.2 now offers Remote direct memory access (RDMA) support for applications and clients with NFS over RDMA. The key benefits of this feature include significantly higher throughput performance, especially for single connection and read intensive workloads. In addition, customers can achieve high bandwidth, low latency communication, and improved cluster utilization for GPU powered applications.

With OneFS we have always seamlessly migrated between generations of hardware. OneFS can grow the namespace and filesystem dynamically and automatically. This provides users and applications with uninterrupted access to the data while nodes and capacity are being added to the cluster. This is equally important for retiring nodes and for in-place migrations, since OneFS can shrink the namespace and filesystem, again without disruption to users and applications.

From the time data is created, to the time it is deleted, data moves through phases of its lifecycle. Now, there is no standard definition for these phases but generally once stored, data must be managed and prepared for usage. It must be secured and protected against exfiltration. Data must be analyzed to uncover actionable insights and opportunities. It needs to be moveable between platforms, locations, and storage tiers. Dell EMC have a growing ecosystem of data management software that can help you answer yes to all these questions below and much more.

• How much data do I have & where is it?
• Is my data secure?
• Is it stored in the right place & accessible?
• Can my data be moved?
• Can I automate processes?

The OneFS 9.2 documentation, can be found, by clicking the screenshot below

Lastly, some additional coverage on the new, FS900 and OneFS 9.2