Warning: Illegal string offset 'url' in /var/www/dataversity/wp-content/plugins/dv-promo/dv-promo.php on line 134
The shift to the Cloud has already had a huge impact on enterprise IT. And more is yet to come. Soon enough, Data Protection will be viewed as a service that enterprise IT will expect from their Cloud service providers in order to deliver data availability for their users. Gartner, for instance, has predicted that the Disaster Recovery as a Service (DRaaS) market is estimated at approximately $2.01 billion, with an expected growth to $3.7 billion through 2021.
Data Protection is becoming the most common use case for Enterprise Hybrid Clouds, said Rich Petersen, President and Co-founder of JetStream Software. The company plans to be a big part of the transition, adding Data Protection services for SaaS, IaaS, and PaaS infrastructure providers that rely on VMware to provide hosted services to their own customers.
The recently announced JetStream Data Protection solution is the latest addition to the company’s Cross-Cloud Data Management Platform, which also comprises JetStream Accelerate and JetStream Migrate.
The newcomer to the portfolio is designed to get data from the data center to the Cloud as efficiently as possible for Continuous Data Protection, Business Continuity, and Disaster Recovery without snapshots. It draws upon capabilities already embedded in JetStream Accelerate, which is designed to optimize data efficiency at scale and whose benefits include increased application performance, improved virtual machine density, and enhanced storage efficiency for data centers and Cloud services.
JetStream Accelerate Origins and Purpose
JetStream Accelerate is the product of a partnership with VMware to co-design the vSphere API for IO Filtering. JetStream Accelerate increases computing and storage efficiency by optimizing storage IO through low-latency tiering in non-volatile memory. IO Filtering itself is a VMware API framework that lets third-party technology be integrated into vSphere for data services.
An IO Filter is a better option for data capture than other solutions, such as snapshots, said Petersen, as the latter slows down application performance, or putting virtual appliances in the data path, which also impacts application performance.
In the VMware environment, the bottleneck on performance usually isn’t the CPU but storage I/O, Petersen explains. JetStream Accelerator is placed on a solid-state device in the host server to overcome the bottleneck of using external, shared hard disk drive storage for data access.
Instead of memory resources sitting idle, waiting for data, “we serve it from flash memory in the host, so you get better application performance. Also, you can put more virtual machines on each individual host server,” he said. That matters a lot to Cloud service providers, who want to maximize their capabilities without raising their costs.
According to the company, JetStream Accelerate software in testing increased the read IOPS of a varying number of virtual machines from 2.5x to 5x compared to HD storage. And one customer, Petersen said, observed an enormous reduction of traffic on underlying storage. Its storage costs are going down because it doesn’t need to over-provision storage.
The recently announced JetStream Accelerate 4.5 for VMware vSphere has the ability to run in application cluster mode for Oracle RAC, in which a single virtual disk is shared by multiple VMs that need to cooperate and communicate to maintain data consistency. This has been implemented at a major European telecommunications and Cloud service provider, and a reference architecture will be published in conjunction with VMware and Oracle about how the environment works.
Data Protection as a Priority
JetStream Accelerate capabilities make their way into JetStream Data Protection: Continuous data capture through IO filter software components deployed to each VMware host avoids the use of software agents in virtual machines and virtual appliances in the data path as part of a VMware Ready solution.
“While data is being written to on-premises storage, it also is written to a low-latency replication log that can be deployed on a non-volatile memory appliance for minimal impact on application performance,” said Petersen.
Its flash-based acceleration tier to the Cloud supports JetStream Data Protection’s guaranteed low recovery point objective and the ability to make sure that a flood of data doesn’t overwhelm a Cloud service’s data center resources. “Cloud data center operators want to manage storage efficiently to keep infrastructure costs low,” Petersen said. And enterprise data center administrators can rely on a policy engine to direct where data is replicated to and recovered.
Importantly, continuous replication of data assures point-in-time recovery. “That’s most needed with malware,” Petersen said. “You can roll back to a point in time before an intrusion took place and recover data from there.”
All this plays a part in what Petersen characterizes as the company’s vision for Cloud-based Data Protection that lets Cloud services providers give their enterprise customers the level of Data Protection they have enjoyed from traditional on-premises Data Protection solutions, with high availability, low latency, and near-instant recoverability.
Taking on Cloud Migration
Shortly before the announcement of JetStream Data Protection, the company’s JetStream Migrate solution became generally available.
It solves one basic problem that Cloud service providers increasingly have. Customers who have virtual machines in their data centers want to move their VMs to managed service providers as they undertake infrastructure refreshes. “But they can’t shut down their VMs for a week while moving data over,” Petersen said. That’s what’s called a “live migration problem.”
“It’s the only solution for Cloud live migration based on IO filters,” said Petersen. The light footprint IO filter is managed by vSphere on-premises and gives the ability for companies to migrate to the Cloud without shutting down VMs for the duration. Said Petersen, “End customers and Cloud service providers are looking for better technologies to do this.”
Image used under license from Shutterstock.com