This week’s News Bits we look at a number of small announcements, small in terms of the content, not the impact they have. StorONE partners with SQream. Samsung mass produces eUFS 3.1. FalconStor launches StorSafe. Quantum completes acquisition of ActiveScale. Microsoft Azure NVv4 Virtual Machines hit general availability. AWS adds PyTorch support. Dynatrace offers extended free trials.
This week’s News Bits we look at a number of small announcements, small in terms of the content, not the impact they have. StorONE partners with SQream. Samsung mass produces eUFS 3.1. FalconStor launches StorSafe. Quantum completes acquisition of ActiveScale. Microsoft Azure NVv4 Virtual Machines hit general availability. AWS adds PyTorch support. Dynatrace offers extended free trials.
StorONE Partners With SQream
StorONE has partnered with SQream to provide users with high-performance massive data analytics. According to the companies, with StorONE Enterprise Storage Platform, S1, SQream DB can saturate a multi-node, parallel file over 100GbE, ensuring the NVIDIA Tesla V100 GPUs the software uses are fully utilized.
Samsung Mass Produces eUFS 3.1
Samsung announced that it has begun mass production on its 512-gigabyte (GB) eUFS (embedded Universal Flash Storage) 3.1 for use in flagship smartphones. The new storage is said to deliver nearly 3x the write speeds of the previous version at about 1.2GB/s. It also offers 100K IOPS read and 70K IOPS writes. This brings storage performance normally thought of in a notebook to a mobile phone.
FalconStor Launches StorSafe
FalconStor Software, Inc. launched StorSafe, what it is calling the industry’s first enterprise-class persistent data storage container. Some of StorSafe’s benefits include:
- Portability across any S3 cloud, object storage, and on-premises storage environments and future-proofing for technology advancements
- Dramatically improved long-term archive accessibility and reinstatement for strategic archive leverage
- Variable payload container storage, allowing hyper-efficient data deduplication over large storage containers to significantly reduce cost and storage capacity consumption, and smaller storage containers optimized for accelerated data reinstatement
- Increased security and data integrity capabilities to maintain the confidentiality, integrity, availability, nonrepudiation, and accessibility of archive data, as well as a journaled integrity checks at user set intervals for data integrity validation, cyber intrusion detection, and verification of chain of custody
- Critical retention data is no longer tied to or integrated with a specific hardware storage system or cloud platform allowing independent data portability over extended retention periods without expensive cross-platform service migration costs
- Redundant Array of Independent Clouds (RAIC) container distribution via multi-cloud erasure coding for data redundancy, redundancy overhead reduction, and accelerated disaster recovery, which delivers 99.99999+ of availability even if an entire cloud provider goes dark
Quantum Completes Acquisition Of ActiveScale
Quantum has completed its acquisition of ActiveScale from Western Digital. The company states that this acquisition will expand its leadership role in storing and managing video and other unstructured data using a software-defined approach.
Microsoft Azure NVv4 Virtual Machines Hit GA
Microsoft and AMD continued to expand their partnership with the announcement of Microsoft Azure NVv4 Virtual Machines, powered by 2nd Gen AMD EPYC Processors and AMD Radeon Instinct GPUs. VMs include:
- Dav4 series VM: The Dav4 and Dasv4 Azure VMs are made for a variety of general-purpose applications. Featuring the powerful AMD EPYC 7452 processor, the VMs offer up to 96 vCPUs, 384 GBs of RAM, and 2,400 GBs of SSD-based temporary storage and support for Azure Premium SSDs.
- Eav4 series VM: The Eav4 and Easv4 Azure VMs are made for memory-intensive workloads. These new VMs were the first in the cloud to feature the AMD EPYC 7452 processor and offer up to 64 percent better SQL Server workload performance in Azure than the previous generation E-series VM
- HBv2 VM: Powered by the AMD EPYC 7742 CPU, these VMs are purpose made for high-performance computing workloads like CFD, explicit finite element analysis, seismic processing, reservoir modeling, rendering and more. Recently, Azure announced that in a series of HPC benchmarks, the HBv2 VM eclipsed 80,000 cores for message passing interface scalability, providing on-premise supercomputing levels of performance in the cloud.
- NVv4 VM: Powered by 2nd Gen AMD EPYC CPUs and AMD Radeon Instinct MI25 GPUs, NVv4 delivers a modern desktop and workstation experience in the cloud. Single Root I/O Virtualization (SR-IOV) based GPU partitioning offers four resource-balanced configuration options, from 1/8th to a full GPU, to deliver a flexible, GPU-enabled virtual desktop experience.
- Lsv2: The Lsv2-series is well suited to big data applications, SQL and NoSQL databases, data warehousing, and large transactional databases. The Lsv2 VMs run on the AMD EPYC 7551 processor.
AWS adds PyTorch Support for Elastic Inference
AWS announced that it added support for PyTorch models with Amazon Elastic Inference (EI). AWS states that this will bring some great benefits to ML developers including: allowing them to only use the required compute power needed, thus reducing the cost of running deep learning inference by up to 75%. Elastic Inference supports TorchScript compiled models on PyTorch in regions where EI is available.
Dynatrace Offers Extended Free Trials
With the Covid-19 pandemic happening, many companies are extending free offerings to boost remote workers. Dynatrace is no different with free trial access to its Software Intelligence Platform, through May 19, 2020. The company states that this will help with the following:
- Ensuring applications and infrastructure run smoothly amid increased demand
- Delivering instant answers on performance problems and business metrics for remote teams
- Enabling remote teams to identify and focus on their highest priority work
