Pure Storage ActiveCluster delivers “Simple Stretch Clustering For All”. This is great news for our customers and is quite a departure from legacy business continuity solutions which are complex and expensive.
ActiveCluster changes everything:
- Active/active design enables I/O to be served on a given volume from two FlashArrays simultaneously
- Failovers are safe, automatic, and transparent
- Pure1® Cloud Mediator, our SaaS-based quorum witness, means no third site to manage, and no extra hardware or software licenses
- Set up is measured in minutes rather than days, and involves just 4 steps and 1 new command;
- Management is at the array-level via Purity Protect; and,
- There are no additional licenses or fees – it is included now in Purity//FA 5 release
Industry analyst firm Enterprise Strategy Group (ESG) recently published a Lab Validation report on Pure Storage ActiveCluster and concluded that it “…extends the opportunity for synchronous replication to any application, and to organizations that don’t have the resources—financial or staffing—that synchronous replication has required in the past.”
The combination of synchronous replication with transparent failover delivers RPO zero, RTO zero for always-on availability. ActiveCluster will tolerate an application host, an array, a datacenter, or even a network failure, ensuring continuous operation of your business enabling the highest level of business continuity for all.
ActiveCluster Challenges The Status Quo
As with everything Pure, ActiveCluster’s architecture exhibits Pure’s core operational, design, and ownership principles: Effortlessness, Efficiency, and Evergreen™. ESG notes that “ActiveCluster makes maximizing uptime simple and cost effective for any application and organization.”
ActiveCluster is a feature available in Purity//FA 5. For existing FlashArray customers with an Evergreen™ subscription, ActiveCluster is available via a simple non-disruptive upgrade. Implementing ActiveCluster is incredibly simple, requiring just four steps and only one new command. CustomerS can be up and running with ActiveCluster in minutes, nevermind days, and certainly not weeks or months.
In contrast, installing and configuring our competitors’ stretch cluster solutions is often complicated, requiring extensive design, preparation, professional services, and time. Each of these competing stretch cluster solutions requires specialized business continuity knowledge and skills. Purity ActiveCluster makes tier-1 enterprise-class business continuity effortless to the storage admin, virtualization admin and even IT generalist, allowing them to focus on helping the enterprise stay productive, rather than implementing, managing, tweaking, and maintaining the array.
ActiveCluster was designed so operations like volume, snapshot and protection management can all be performed using the GUI, CLI, or REST API in a way customers are already familiar with. via aManaging an ActiveCluster stretch cluster is as simple as managing a single FlashArray.
Leveraging the Pure1 Cloud Mediator, you do not have to additionally configure and provision a separate 3rd site of your own, failovers are completely transparent to hosts and applications, and the arrays automatically re-synchronize. The Pure1 Cloud Mediator is unique among in-market stretch cluster solutions. Pure1 Cloud Mediator affords a simple to use, lightweight, and automatic way for applications to transparently failover, or simply move, between sites in the event of a failure without any user intervention.
It has numerous advantages over traditionally implemented quorum witnesses, including being SaaS-based (i.e. there’s nothing to install onsite, no hardware or software to maintain, nothing to configure and support for HA, no security patch updates, etc.), it’s inherently in a separate failure domain from either of the two FlashArrays, and it requires no human intervention.
In contrast, Dell EMC, HPE, and NetApp stretch clusters require an incremental 3rd-site for a (typically VM-based) quorum witness to operate. While these competitor-required 3rd-sites can be on-premises or cloud-based, they can also require extra compute and storage, and software licenses. Note that, in addition to a required quorum witness, a Dell EMC VPLEX/Metro stretch cluster is more complex: each of the two VPLEX clusters (one per site) requires one VPLEX Management Console and at least one (up to 4) VPLEX Engine(s). ActiveCluster entirely precludes these extraneous site, hardware, and software requirements.
ActiveCluster also affords effortless non-disruptive volume migration: on a source FlashArray. Customers can choose to add volumes to a pod (effectively, a replication group), then stretch the pod to encompass a second FlashArray, let it synchronize, and then unstretch the pod. This moves the chosen volume to a target FlashArray simply, with no fuss.
Dell EMC supports non-disruptive data migration with VPLEX/Metro – called global migration. Their documentation states that VPLEX/Metro migrations, which employ large data transfer sizes, incur more performance impact on inter-cluster I/O. However, small transfer sizes result in more notification broadcasts, thereby slowing the data migration.
Specific to the HYPERMAX OS for VMAX storage arrays, Dell EMC states that “[SRDF] Devices [(aka Symmetrix volumes)] that are part of an SRDF/Metro configuration cannot be migrated.”
Live volume migration with HPE’s 3PAR Peer Persistence requires that “…the standby paths are marked as active and host traffic to the primary storage array is redirected to the secondary storage array without impact to the hosts.“
These are examples of the kinds of legacy storage issues Pure Storage administrators do not have to worry about.
Customers have great confidence in the ability of ActiveCluster to support business continuity, in part because it is easy to test. Simple to perform and straight forward testing is a key benefit of ActveCluster. Because of its efficient design, customers can simulate multiple failures of varying severity easily – a total site failure, an array failure, a network failure, etc.
In ESG’s Lab Validation of ActiveCluster, “Many organizations set up business continuity and disaster recovery solutions, but never test them because of complexity and disruption to production operations. And, sadly, many find the problems and configuration issues only during a disaster, when they cannot access their data. Purity ActiveCluster takes the cost and complexity barriers out of business continuity, making it simple, cost-effective, and provable for any organization—and relieving the minds of IT administrators, auditors, and other compliance officers.”
The Purity Operating Environment (Purity OE) delivers purpose-built, comprehensive, tier-1 data services to all workloads, natively. With Purity Protect, FlashArray snapshots and replication are built-in, data reduction-aware, and simple to set up. Complete policy-based automation ensures that you get protection within minutes. Included with Purity//FA 5, ActiveCluster is enabled by the native replication and snapshot capabilities of Purity Protect.
ActiveCluster benefits from the Purity Reduce data reduction capabilities. Pure delivers an average ~5:1 data reduction ratio (i.e. deduplication and compression only), with a greater than 10:1 total efficiency ratio across mixed-workloads on the FlashArray installed-base.
Known as Purity Assure, FlashArray resiliency has delivered tier-1 caliber, enterprise-class proven greater than 99.9999% data availability, inclusive of maintenance, failures, and generational upgrades. This level of reliability and availability is proven across our FlashArray installed base doesn’t count the increase in availability afforded by ActiveCluster.
On the contrary, Dell EMC states that six 9s data availability on the VMAX All-Flash series requires the addition of SRDF/A, their asynchronous replication offering (and a second VMAX array).
As with Pure’s entire portfolio, ActiveCluster is delivered via our industry pioneering and customer-centric ownership model – the Evergreen™ Subscription to innovation. Customers who are subscribed to Evergreen have access to our future innovations as they become available – whether it is hardware or software. This is because Evergreen is far more than a set of disparate sales and marketing programs, but the underlying design principle of Pure is that customers should not have to repurchase, relicense, or migrate data to gain access to our innovations.
Evergreen affords FlashArray customers access to software updates and new features, both now and in the future. In contrast, for competing stretch cluster offerings: Dell EMC offers VPLEX as a separately sold and expensive offering for XtremIO X2 (requiring the optional AppSync software) and Unity All Flash series, while SRDF/Metro is ‘included’ only in the optional and extra-cost “FX” software package for its VMAX All-Flash series.
HPE’s 3PAR Peer Persistence requires “…a frame or array-level license and is required to be licensed once per the lifetime of any array that is participating in a replication or federation group.”
The experience Purity ActiveCluster provides is simple and can be trusted to provide business continuity at no additional cost, which is quite different from our competition.
The best proof to support this is what customers say themselves. Check out two customer blogs on ActiveCluster and the benefits it provides enterprises of all sizes.
- ESET uses ActiveCluster for simplified business continuity and their Solution Architect, Brian Cummings, had a wide variety of experiences with competitor solutions prior to using Pure. “Over the years, we have tried many different solutions for business continuity, and none has been as remarkable as ActiveCluster from Pure Storage. ActiveCluster performs synchronous replication between remote sites automatically and transparently, without downtime, human intervention or added cost.”
- Kremsmueller commends the ease of implementation and the high value ActiveCluster provides their organization. “Business continuity is critical for all businesses of any significant scale and Pure’s ActiveCluster is an easy solution that solves a historically complex and expensive problem.” said Alexander Ietan, IT infrastructure leader at Kresmueller.
Learn more by reading the ESG Lab Validation Report of Pure Storage ActiveCluster.
For a detailed look at implementing ActiveCluster, please visit Setting Up FlashArray Active/Active Replication (ActiveCluster).