March 8, 2016

Everspin on Wednesday introduced the nvNITRO storage accelerators that deliver extremely fast read/write times with low latency. The initial capacities of the cards are 1GB and 2GB of Spin Torque MRAM 256Mb DDR3 ST-MRAM.

The nvNITRO ES1GB and ES2GB operate at 1.5 million IOS with 6 microsecond end-to-end latency. Configured in a half-height, half-length form factor, the cards have two nodes: NVMe SSD and memory mapped IO (MMIO). 

Why Should You Care: These new PCIe cards communicate in two modes, one for NVMe SSDs and the other for memory mapped IO (MMIO).  This means these new persistent memory technologies have a PCIe-based accelerator card to help facilitate performance specifically for these technologies.

March 8, 2017

Excelero, a software-defined block storage vendor, on Wednesday announced version 1.1 of its NVMesh Server SAN software. NVMesh is an NVMe sharing technology that allows customers to create virtual, distributed non-volatile arrays.

Excelero uses a patented Remote Direct Drive Access (RDDA) technology in its NVMesh that allows the logical disaggregation of storage from application servers, keeping CPU available for applications.

It adds 5 milliseconds latency over local NVMe drive latency and supports NVMe over Fabrics.

Why Should You Care: Excelero is coming out of stealth and putting a twist on SDS for block storage. This software is step closer to communicating with flash and other solid state technologies such as persistent memory instead of traditional HDD deployments.  This means users can now take advantage of the full performance and capabilities of NVMe, NVMe over Fabric and new memory technologies, such as 3D XPoint, in the near future to increase productivity at all levels of business.

 

 

March 7, 2017

HPE on Tuesday announced that it has acquired Nimble Storage for $1.0 billion cash to bolster its hybrid IT platforms.

The acquisition, priced at $12.50 per share, is expected to close at $1 billion. Nimble plays in the entry- to midrange storage segment and its offerings are complementary to its 3PAR and MSA products.

The acquisition will allow HPE to offer flash storage across segments. HPE also plans to incorporate Nimble’s InfoSight Predictive Analytics across it storage portfolio.

Why Should You Care: This acquisition clearly closes a gap in HPE’s range of products.  HPE seems to want to not only fill a gap in their portfolio for entry level flash but also add the predictive analytics that Nimble products bring.  This means HPE customers now have additional products to choose from to fit their needs.

 

 

March 6, 2017

SwiftStack on Monday announced a new version of its latest hybrid cloud storage and data management package, SwiftStack 5, which now includes the ability to replicate data bidirectionally between four Google Cloud Storage offerings.

Two-way synchronization of data stored between on-premises and the public cloud is managed as a policy controlled by IT, so data is automatically delivered to where it can best be computed on or utilized, whether in private data centers or in the public cloud.

SwiftStack 5 stores data in a cloud-native format and manages its placement via policies. Other enhancements include multi-region erasure coding for improved storage efficiency as well as support for deep buckets with more than 100 million objects.

The SwiftStack 5 release includes a new desktop client for Windows and OS X.

Why Should You Care: These enhancements give customers in hybrid cloud environments the ability to manage both on-premises and off-premises data policies with Cloud Sync. This will make data management easier in these environments.

 

March 1, 2017

 HubStor last week added security and shadow copy capability to its HubStor hybrid cloud storage and archive service.

HubStor customers can now protect against tampering at the account or subscription level and shadow their data to a secondary account as a backup.

With HubStor’s shadow copy option, malicious users are unaware that a shadow copy exists. HubStor transparently maintains replicas in a secondary account during the synchronization and archiving process.

Why Should You Care: HubStor now offers synchronous replication of data to a secondary location for data protection. This secondary copy gets stored as a shadow copy.  This means additional data protection for users not only from data loss, but due to account deletion. It can also be used for disaster recovery.

 

March 5, 2017

Cisco and IBM last week announced new VersaStack converged infrastructure appliances that include hybrid cloud capabilities and added software defined storage capability.

VersaStack now uses Cisco CloudCenter and IBM Spectrum Copy Data Management to deploy, provision, and manage applications and data and now supports 20-plus cloud types and regions. It also leverages Spectrum Software Define Storage to improve the efficiency of existing server and storage resources. Finally, VersaStack now supports scaling to multi-petabyte storage via Cisco UCS S-Series Storage Servers and IBM Cloud Object Storage.

Why Should You Care: With their competitors already in the converged infrastructure business, this announcement is to keep up with the competition.  This means there is yet another converged infrastructure solution on the market to reduce complexity and increase rapid deployment.

 

March 6, 2017

IBM on Monday announced the first commercially available initiative to build commercially available quantum computers. Called IBM Q, quantum systems and services will be delivered and available via the IBM Cloud.

The company also introduced a new API for its quantum computer that allows developers to start building interfaces between its existing five quantum bit (qubit) cloud-base quantum and classic computers.

The release of an upgraded simulator on the IBM Quantum Experience that can model circuits with up to 20 qubits. In the first half of 2017, IBM plans to release a full SDK (Software Development Kit) on the IBM Quantum Experience for users to build simple quantum applications and software programs.

Why Should You Care: Unlike traditional computers that can solve problems by using existing datasets to see patterns, IBM Q is based on quantum physics. This means solving problems with data that doesn’t exist and using probability to solve problems traditional computers can’t process.