Monday, July 9, 2018

IHS Markit: Flourishing on-prem enterprise data center market

The on-premise, enterprise data center market is flourishing, according to the newly released Data Center Strategies and Leadership North American Enterprise Survey from IHS Markit.

“Application architectures are evolving with the increased adoption of software containers and micro-services coupled with a Dev/Ops culture of rapid and frequent software builds. In addition, we see new technologies such as artificial intelligence (AI) and machine learning (ML) incorporated into applications. These applications consume network bandwidth in a very dynamic and unpredictable manner and make new demands on servers for increased parallel computation,” said Cliff Grossner Ph.D., senior research director and advisor for cloud and data center at IHS Markit.

Some highlights of the report:


  • Respondents expect a greater than 2x increase in the average number of physical servers in their DCs by 2019.
  • Top DC investment drivers are security and application performance (75% of respondents) and scalability (71%).
  • 9% of servers are expected to be 1-socket by 2019, up from 3% now.
  • 73% of servers are expected to be running hypervisors or containers by 2019, up from 70% now.
  • Top DC fabric features are high speed (68% of respondents), automated VM movement (62%), and support for network virtualization protocols (62%).
  • 53% of respondents intend to increase investment in software-defined storage, 52% in NAS, and 42% in SSD.
  • 30% of respondents indicated they are running general purpose IT applications, 22% are running productivity applications such as Microsoft Office, and 18% are running collaboration tools such as email, SharePoint, and unified communications in their data centers.
  • Cisco, Dell, HPE, Juniper, and Huawei were identified as the top 5 DC Ethernet switch vendors by respondents ranking the top 3 vendors in each of 8 selection criteria.


Facebook presentation: Optics Inside the Data Center

Mark McKillop, Network Engineer at Facebook, and Katharine Schmidtke, Sourcing Manager of Network Hardware at Facebook, talk about challenges in Facebook's optical networks, both in backbone and in data centers.

The first part of the video covers the optical systems used to connect Facebook's POPs and data centers.

The second part discusses optical scaling challenges inside the data centers, including the potential for onboard optics in future systems.

This 30-minute video presentation was recorded at Facebook's Networking@Scale 2018 event in June in California.

See video:
https://www.facebook.com/atscaleevents/videos/2090069407932819/

Facebook's announced data centers:

U.S
Prineville, Oregon
Los Lunas, New Mexico
Papillon, Nebraska
Fort Worth, Texas
Altoona, Iowa
New Albany, Ohio
Henrico County, Virginia
Forest City, North Carolina
Newton County, Georgia

Europe
Clonnee, Ireland
Lulea, Sweden
Odense, Denmark



Facebook looks to Spiral for self-tuning, real-time services

Facebook's engineering group is developing a small, embedded C++ library called Spiral to providing the self-tuning of its data infrastructure necessary to fine-tune thousands of micro services.
Machine learning is used with Spiral to create data-driven and reactive heuristics for resource-constrained real-time services.

Facebook says software maintenance increasingly requires self-tuning capabilities because it is simply "too difficult to rewrite caching/admission/eviction policies and other manually tuned heuristics by hand."

https://code.fb.com/data-infrastructure/spiral-self-tuning-services-via-real-time-machine-learning/


Samsung begins production of 5th gen 256Gb V-NAND memory

Samsung Electronics has begun mass producing its fifth-generation V-NAND memory chips.

The new memory is the first to use the "Toggle DDR 4.0" interface, which offers a data transfer speed of 1.4 Gbps between storage and memory -- a 40% increase from Samsung's 64-layer predecessor. Samsung said its new V-NAND also has the fastest data write speed to date at 500-microseconds (μs), which represents about a 30% improvement over the write speed of the previous generation, while the response time to read-signals has been significantly reduced to 50μs.

Samsung’s fifth-generation V-NAND packs more than 90 layers of ‘3D charge trap flash (CTF) cells,’ the largest amount in the industry, stacked in a pyramid structure with microscopic channel holes vertically drilled throughout. These channel holes, which are only a few hundred-nanometers (nm)-wide, contain more than 85 billion CTF cells that can store three bits of data each. This state-of-the-art memory fabrication is the result of several breakthroughs that include advanced circuit designs and new process technologies.

“Samsung’s fifth-generation V-NAND products and solutions will deliver the most advanced NAND in the rapidly growing premium memory market,” said Kye Hyun Kyung, executive vice president of Flash Product and Technology at Samsung Electronics. “In addition to the leading-edge advances we are announcing today, we are preparing to introduce 1-terabit (Tb) and QLC (quad-level cell) offerings to our V-NAND lineup that will continue to drive momentum for next-generation NAND memory solutions throughout the global market.”

Radisys releases decomposed Virtual Media Server for NFV

Radisys released a decomposed Virtual Media Server for NFV that enables service providers to scale control and media elements independently. Decomposing the architecture allows application developers and service providers to support tens of millions of subscribers through application servers that leverage a large pool of media server resources for services such as conferencing, speech recognition, and real-time video.

Radisys said its MediaEngine Virtual Media Server simplifies application development and deployment, significantly lowering cost of service delivery and accelerating time-to-market for new services.

Key new capabilities include:

  • Support for new NFV-ready management solutions such as ONAP and OpenStack. This enables service providers to deploy media applications on demand in a network – for example delivering edge media for real-time video to devices in a concert or sports arena, with network resources being able to be re-deployed once the event is over, all while keeping the traffic local to minimize end-to-end network bandwidth consumption.
  • New advanced revenue-generating media applications, such as deep media analytics and reporting for on-demand media optimizations, and for integrated speech recognition for speech-enabled applications.
  • Hardware accelerated virtualized real-time voice and video services using Web/4G/LTE networks and advanced media optimizations at the edge using Multi-Access Edge Computing platforms.
  • NFV-compliant MediaEngine Management Software to simplify multi-node MediaEngine management deployed in public and private clouds.

“The evolution of the MediaEngine Virtual Media Server to enable on-demand media anywhere in the network is transformational in matching the dynamic needs of service providers and their customers with the industry’s leading economics for one of the most demanding functions in carrier and cloud networks,” said Al Balasco, vice president, MediaEngine, Radisys. “We’re enabling our customers to achieve massive savings in CapEx and OpEx when deploying the Virtual Media Server to support millions of subscribers in a server footprint that is typically half to one-tenth the footprint of alternative solutions. Not only does this reduce costs for existing services, it also makes economically viable a range of previously cost-prohibitive services, further expanding service providers’ ability to offer new services and differentiate in the market.”


Google Cloud adds cost forecast tool

Google Cloud Platform (GCP) introduced a cost monitoring and forecasting tool to help customers manage their bill.

The tool tracks all services and projects managed inside of GCP.

https://cloudplatform.googleblog.com/


Seaborn adds Mehmet Akcin to Board

Seaborn Networks, a leading developer-owner-operator of submarine fiber optic cable systems, has added Mehmet Akcin to its Board of Advisors.

Mehmet is one of Silicon Valley's most innovative engineering executives, with 15 years of experience and 6 patents granted to his name while working at Microsoft. His areas of expertise include submarine cables, IP backbones, Domain Name Systems, network security and content delivery networks. Until June 2018, he served as Senior Director, Global Infrastructure Planning and Acquisitions group at Oath (formerly known as Yahoo!), a subsidiary of Verizon Communications Inc. Previously, he worked at Microsoft as a Director in the Global Network Acquisition Group and as Chief Engineer at Internet Corporation for Assigned Names and Numbers (ICANN).

See also