JetStream Software Unveils Next-Generation Cross-Cloud Data Protection SoftwareAugust 21, 2018
JetStream Software Inc., an innovator in cloud software development, today announced JetStream Data Protection, enabling cloud service providers an advanced solution to deliver data availability services to enterprise customers. The JetStream Data Protection software joins the already available solutions, JetStream Accelerate and JetStream Migrate, as part of the JetStream Cross-Cloud Data Management Platform.
JetStream Software will be exhibiting at VMworld 2018 in the New Innovators Pavilion, August 26-29, 2018, in Las Vegas.
"Data protection is the most common use case for enterprise hybrid cloud," explained Tom Critser, co-founder and CEO of JetStream Software. "So it is no surprise that the market is moving away from legacy snapshot-based replication architectures and adopting continuous replication methods for cloud-based data resiliency."
Advanced Design for Data Protection Services
The JetStream Data Protection software provides a spectrum of data protection capabilities for an on-premises VMware environment, including:
- Non-disruptive continual data replication to the cloud for recovery and failover
- Virtual machine failover to a cloud-based business continuity service
- Virtual machine backup to cloud-based NFS stores and/or object stores
- Policy-based availability for cost-efficient SLA management per recovery group
- Extreme low-latency logging on non-volatile memory
Continuous Data Capture Through IO Filters
Many traditional backup technologies still capture data in intermittent snapshots, which introduces latency in application performance and limits the minimization of recovery point objectives (RPO). As an alternative to snapshots, some technologies capture data input-output (IO) through software agents or virtual appliances, methods that introduce a number of performance and compatibility problems. Continuously capturing data through an IO Filter does not suffer those drawbacks.
In VMware environments, the JetStream solution enables continuous data capture through the VMware vSphere APIs for IO Filtering (VAIO), ensuring compatibility with the entire VMware environment, including any type of on-premises hardware such as SAN, NAS, Virtual SAN or other hyperconverged infrastructure (HCI). Because it employs the VAIO framework for continuous data capture, the solution doesn’t require software agents in the virtual machines, doesn’t require a virtual appliance in the data path, and is a fully supported VMware Ready solution.
The JetStream Software engineering team collaborated with VMware as the co-design partner for the IO Filter API framework. JetStream Software has already released two products based on the IO Filter APIs: JetStream Accelerate for cloud resource and performance optimization and JetStream Migrate for seamless live migration of VMs to the cloud.
JetStream Cross-Cloud Data Protection
The key components of the JetStream Data Protection software include:
- IO Filter: the software component deployed to each VMware host for continuous data capture for every virtual machine in the protected recovery group
- Data Availability Policy Engine: the management component that defines data replication and instance recovery policies for each recovery group
- Low Latency Replication Log: the mechanism for maintaining the replication log, which can be deployed on a highly available non-volatile memory appliance for minimal impact on application performance
- Cloud Store Adapter: software that pre-processes data for backup to a specific object store or NFS store
- Crash-Consistent Data Pipe: high-throughput, multi-channel software connection that provides a crash-consistent transfer of data from the on-premises data center to the cloud destination, ensuring continuous data coherence
- Cloud Acceleration Tier: an extreme low latency data tier in the cloud that enables both zero RPO with data consistency and efficient use of cloud resources
Why Is This Architecture Important?
JetStream Software co-founder and CTO Serge Shats described the reason for the unique design of the JetStream Data Protection software. "We have the ability to continuously capture data for recovery in real time, as it’s being written to storage, but that’s just the start. Beyond capturing data through the IO Filter, we have to satisfy a considerable number of objectives."
These objectives include:
- No degradation of application performance on premises
- High scalability to support environments with thousands of virtual machines
- Low latency replication logging for scalable continuous data replication
- Support for multiple, concurrent availability policies, assigned per recovery group
- Replication to the recovery destination most appropriate for each policy, from a warm VMware environment for cloud failover to an economical object store backup site
- Crash-consistent connectivity with each recovery destination
- Multiple recovery options: e.g., point-in-time (CDP) or whole group failover (DR)
There are hundreds, sometimes thousands, of VMs across the data center, and they’re constantly writing data. "It’s a firehose of data I/O," explained Shats. "If the connection to the cloud, relatively speaking, looks more like a garden hose, you have a problem. So to provide a highly scalable solution for data replication to the cloud, we employ 1) a low-latency environment for the replication log on premises; 2) the multi-channel, high-throughput data pipe to transport data to the cloud; and 3) a high-performance data acceleration tier in the cloud recovery destination."
Pricing and Availability