Monday, March 14, 2016

ONOS Releases Falcon Software for Service Provider SDN and NFV

The ONOS community announced its latest open source SDN Network Operating System release - Falcon, featuring additional support for Central Office Re-Architected as a Data Center (CORD), many new applications and more support for southbound protocols. This marks the 6th software release since the ONOS project became public in December 2014.

Some highlights of Falcon:

  • ONOS is now a part of the OPNFV distribution called Brahmaputra, thanks to the efforts of Huawei who developed OpenStack Neutron ML2/L3 interfaces/features, VTN, Service Function Chaining, OVSDB and Installers' Plugin. 
  • ONOS continues to be a key component of ONF's Atrium distribution, and Falcon will be integrated with the most recent ONF Atrium 2016/A distribution through the ONOS project's collaboration with the ONF.
  • Continued expansion of the REN networks including new network support with NCTU Taiwan, G√ČANT Europe, KREONET Korea, AARNET Australia, AmLight South America.
  • Troubleshooting applications from FNLab/BUPT china.
  • OSPF southbound protocol support from Huawei and Cognizant and a SNMP southbound provider from BTI.
  • New GUI views including the driver matrix view and the application view.
  • New northbound reservation capabilities from Fujitsu
  • Dynamic cluster scaling support.

"ONOS has rapidly evolved to become a platform for service providers to monetize SDN and NFV, while helping vendors and service providers alike to create SDN and NFV solutions leveraging open source and invent new business models," said Bill Snow, vice president of engineering at ON.Lab. "The ONOS project continues rapid innovation and delivery with quarterly releases, and the capability of each release continues to improve with accelerating participation of the community. It is really exciting to see the new PoCs. Our goal has always been to methodically move the use case ideas from concept to trial and then into production."

http://www.linuxfoundation.org

Pure Storage Scales All-Flash to Multi-Petabytes at <$1/GB

Pure Storage introduced an all-flash storage platform designed to scale to multi-petabyte data sets at economics of less than $1/GB usable.

The company said its new FlashBlade architecture makes all-flash storage affordable for even the largest file and object environments, opening up the performance gains of flash to a wide range of applications.

Pure's new FlashBlade architecture combines commodity hardware components with custom software and ultra-dense packaging. FlashBlade starts at <100 1.6pbs="" 4u="" a="" and="" be="" build="" can="" capacity="" chassis="" effective="" grow="" in="" interconnected="" larger="" multiple="" of="" p="" store="" systems.="" tbs="" to="" up="">
The system three core elements:

  • Blade: The core scale-out element of FlashBlade, each blade contains compute power for data and metadata processing, raw NAND flash for storage and NV-RAM capacity for power-safe protection of writes in flight. Blades are engineered to be ultra-dense and low-power, in 8TB and 52TB capacities.
  • Elasticity Scale-Out Software: Running in a distributed fashion across all blades, Elasticity is the software heart of FlashBlade. Elasticity implements a common object store in the system, storage services such as data reduction, encryption and N+2 erasure coding, scale-out protocol services such as NFS (file) and S3 (object), as well as low-level flash management and LDPC error correction. Because Elasticity is one code base across flash management, protocols, volume and storage services, all layers share a common extensible metadata engine and garbage collection for higher global system efficiency.
  • Elastic Fabric: FlashBlade includes a built-in, software-defined, low-latency 40 Gbps Ethernet network. The Elastic Fabric combines blade-to-blade, chassis interconnect and client connectivity into one network, with software-defined QoS to ensure proper traffic prioritization. Both the Elastic Fabric and Elasticity leverage proprietary protocols on top of the commodity Ethernet network for the highest performance low-latency communication.

Significantly, FlashBlade enables an elastic scale-out, where storage capacity, IO and metadata performance, bandwidth, and client connectivity grow linearly as 8TB or 52TB blades are added non-disruptively to the system. FlashBlade delivers up to 15 GB/s of bandwidth per 4U chassis and consistent all-flash low-latency. The Elasticity OS supports a gigantic scale-out namespace that can be accessed via both file and object protocols, and can adapt to future protocols.

“The value that can be gained from efficiently storing and analyzing unstructured data has a transformative affect on business,” said John Hayes, Co-founder and Chief Architect, Pure Storage. “FlashBlade unlocks new capabilities and use cases that are simply not possible with today’s existing solutions by delivering real-time analytics at massive scale and blazing speeds to enable the future of innovation.”

http://www.purestorage.com

Video: Infinite Storage Bandwidth

We're facing a really big problem in shared storage because CPU bandwidth is not keeping up with network bandwidth or SSD bandwidth, says Fritz Kruger, Engineering Fellow and Chief Hardware Architectu for SanDisk System and Software Solutions.

This video looks at the widening performance gap between network, storage and DRAM bandwidth.

YouTube: https://youtu.be/JHcIVhq199Q




HPE Intros Hyperconverged Appliance

Hewlett Packard Enterprise introduced an all-in-one compute, software-defined storage and intelligent virtualization appliance for mid-sized companies and remote office branch offices.

HPE's new Hyper Converged 380 (HC 380), which is optimized for rapid VM rollout, is built on HPE StoreVirtual VSA. Built in analytics and tools enables IT to reduce VM sprawl and over-provisioning of VMs by 90 percent.

"Hyper-convergence presents a huge opportunity for mid-sized and ROBO enterprises who face the unique challenge of having to stay competitive at enterprise-scale, but with notably less resources," said Ric Lewis, senior vice president and general manager, Converged Datacenter Infrastructure, HPE. "HPE is the only vendor that can offer them the ability to take advantage of the benefits of hyper-converged environments today and provide them with a clear pathway to a composable infrastructure future when they are ready."

http://www.hpe.com

See also