Cisco: Global Cloud Traffic Growing at 35% CAGR

Global data center traffic will triple from 2.6 zettabytes in 2012 to 7.7 zettabytes annually in 2017, representing a 25 percent CAGR, according to a newly published Cisco Global Cloud Index, but the transition to cloud services is driving global cloud traffic at a growth rate greater than global data center traffic. Global data center traffic will grow threefold (a 25 percent CAGR) from 2012 to 2017, while global cloud traffic will grow 4.5-fold (a 35 percent CAGR) over the same period.  Cisco estimates that cloud traffic globally will grow from grow from 46 percent of total data center traffic (98 exabytes per month or 1.2 zettabytes annually) of total data center traffic in 2012 to 69 percent of total data center traffic (443 exabytes per month or 5.3 zettabytes annually) of total data center traffic by 2017.

"People all over the world continue to demand the ability to access personal, business and entertainment content anywhere on any device, and each transaction in a virtualized, cloud environment can cause cascading effects on the network," said Doug Merritt, Senior Vice President, Product and Solutions Marketing, Cisco. "Because of this continuing trend, we are seeing huge increases in the amount of cloud traffic within, between and beyond data centers over the next four years."

Some highlights of the Cisco study:

• Cloud traffic growth by region: The Middle East and Africa will have the highest cloud traffic growth rate from 2012 to 2017 - The Cisco Global Cloud Index now includes regional forecast data for cloud traffic growth.

o   In 2012, North America generated the most cloud traffic (469 exabytes annually), followed by Asia Pacific (319 exabytes annually) and Western Europe (225 exabytes annually).

o   By 2017, North America will generate the most cloud traffic (1.886 zettabytes annually), followed by Asia Pacific (1.876 zettabytes annually) and Western Europe (770 exabytes annually).

• Workload transitions: From 2012 to 2017, data center workloads will grow 2.3-fold; cloud workloads will grow 3.7-fold - In 2012, 39 percent of workloads were processed in the cloud, with 61 percent being handled in a traditional data center. 

• 2014 will be the first year when the majority of workloads shift to the cloud; 51 percent of all workloads will be processed in the cloud versus 49 percent in the traditional IT space.
• By 2017, nearly two-thirds or 63 percent of workloads will be processed by cloud data centers; 37 percent will be processed by traditional data centers.
• The ratio of workloads to nonvirtualized cloud servers will grow from 6.5 in 2012 to 16.7 by 2017. Comparatively, the ratio of workloads to nonvirtualized traditional data center servers will grow from 1.7 in 2012 to 2.3 by 2017.

• Workload growth by region: By 2017, North America will have processed the most cloud workloads, followed by Asia Pacific - The Cisco Global Cloud Index now includes regional forecast data for workload growth.
• In 2012, North America had the most cloud workloads (15.2 million or 47 percent of global cloud workloads), followed by Asia Pacific, which had 6.8 million or 21 percent of global workloads in 2012.
• By 2017, North America will process the most cloud workloads (48.2 million or 41 percent of global cloud workloads), followed by Asia Pacific, which will have 36.5 million or 31 percent of global workloads by 2017.
• From 2012 to 2017, the Middle East and Africa region is expected to have the highest cloud workload growth rate (45 percent CAGR), followed by Asia Pacific (40 percent CAGR) and Central and Eastern Europe (31 percent CAGR).
• Globally, except in the Middle East and Africa and Asia Pacific, traditional data center workloads (rather than cloud workloads) will grow at a low-single-digit CAGR from 2012 to 2017.
• http://newsroom.cisco.com/press-release-content?type=webcontent&articleId=1274405

Read More

Broadcom Unveils its ARMv8-A Multi-Core Processor Architecture for NFV

Broadcom unveiled its ARMv8-A architecture for a new generation of multicore processors designed to deliver server-class performance for network functions virtualisation (NFV).

The new design is built around a 64-bit ARM core with virtualized accelerators for networking, communications, big data, storage and security applications. The design also features a quad-issue, quad-threaded 64-bit ARMv8-A core with superscalar out-of-order execution. The core enables 3-GHz performance in the advanced 16-nm FINFET process node.

“By offering the industry’s highest performance ARM-based multi-core processor architecture, Broadcom is expanding its technical leadership in multiple generations of multi-core processing,” said Ron Jankov, Broadcom Senior Vice President and General Manager, Processor and Wireless Infrastructure. “Our innovations in the CPU core and comprehensive virtualization, along with our adoption of 16-nm FINFET technology, will further separate us from our competitors.”

Broadcom and ARM announced a partnership to define and develop an open, standards-based NFV software environment for the ARM ecosystem. The companies said they will work with members of the Linaro Networking Group and European Telecommunications Standards Institute (ETSI) to standardize the environment’s programming model, tool chains, application programming model, tool chains, application programming interfaces and networking-specific libraries across the industry.

“ARM and Broadcom have enjoyed a close and long-standing partnership. We are now pleased to see Broadcom leverage the ARMv8-A architecture and push the envelope with a very high-performance processing solution for the communications market,” said Tom Cronk, executive vice president and general manager, Processor Division, ARM. “Broadcom is taking a leadership role in helping us to drive an open, standards-based NFV software environment. This environment will enable the efficient and intelligent technology needed by businesses and consumers in our increasingly connected and data-hungry world.”

Broadcom said its goal is to enable an open, portable software environment that enables developers to seamlessly migrate designs based on Broadcom’s XLP II family of multi-core processors to the new ARMv8-A processor architecture, while preserving software investment.

http://www.broadcom.com

In October 2012, Broadcom began sampling its 28nm XLP 200-Series network processor for enterprise, service provider 4G/LTE, data center, cloud computing and software defined networking (SDN) equipment.  The processor family, which is the world's first 28nm multicore communications processor family, promises up to 400 percent faster performance than competing solutions while lowering power consumption by up to 60 percent. 

Broadcom said the product launch demonstrates its successful integration of NetLogic Microsystems, which developed the XLP processors.

The new processors combine quad issue, quad threading and 2 GHz out-of-order execution capabilities with integrated networking and security acceleration.  The XLP 200-Series is the first to integrate a grammar processing engine, a fourth generation regular expression (RegEx) engine, and a broad range of autonomous encryption and authentication processing engines to deliver comprehensive Layer 7 deep-packet inspection (DPI) capabilities and complete offload of the compute-intensive security functions from the CPU cores. 

Read More

Infinera. Brocade and ESnet Collaborate on 100G Multilayer SDN

Infinera and Brocade, in collaboration with the U.S. Department of Energy’s Energy Sciences Network (ESnet), have demonstrated how SDN can be used to automatically provision services and optimize network resources, such as dynamically increasing or rerouting data center to data center interconnect bandwidth services, across a multi-layer network as traffic demands change.

About the demo:

• The first phase of this new demonstration leveraged ESnet’s On-Demand Secure Circuits and Advance Reservation System (OSCARS) in conjunction with an open source controller to provision services over network elements at both the packet and transport layers via the OpenFlow protocol. A 100 GbE service was provisioned across the multi-layer network comprising of Brocade MLXe Core Routers and additional OpenFlow-enabled Layer 2 switches connected over a virtualized transport network consisting of three Infinera DTN-X platforms, each running Infinera’s Open Transport Switch software module. Multi-layer provisioning using SDN enables services and bandwidth to be dynamically provisioned via a single screen across the router layer and transport layer to speed service delivery and save operational costs.
• The second phase of the demonstration focused on multi-layer network optimization and demonstrated how proactive detection of increasing bandwidth of packet flows could trigger OSCARS to dynamically re-route the large flow to bypass an intermediate packet switch/router. This approach leverages SDN to maximize the use of the intelligent DTN-X transport layer with integrated switching to minimize transit traffic carried at the router layer. Automated multi-layer optimization analyzes network topology and traffic flows to route traffic at the most cost-effective and energy-efficient layer of the network to save both capital and operational costs.
• OSCARS contains algorithms for multi-layer provisioning, topology modeling and path computation that have been developed as part of DOE-funded network research. With OSCARS leveraging Floodlight's RESTful application programming interfaces (API), the open source controller is able to discover and provision network elements at both the packet and transport layers via the OpenFlow protocol. Infinera’s DTN-X is a transport platform that converges DWDM and OTN switching without compromise and, in conjunction with GMPLS intelligence and the OTS software, is an optimal platform for presenting a high capacity and flexible virtualized transport topology to SDN controllers such as Floodlight or OpenDaylight. The terabit scale Brocade MLXe routers are fully programmable and offer the industry’s only OpenFlow Hybrid Port Mode with hardware support for OpenFlow L2+L3 12-tuple matching at line rates up to 100 GbE speeds and industry-leading support for up to 128K flows.

“This pioneering demonstration showcases the value of an Intelligent Transport Network combined with SDN control,” said Dave Welch, President, Infinera. “Our converged DTN-X platform is optimal for realizing many SDN benefits, including multi-layer optimization which increases the efficiency of network resources used at higher layers of the network.”

“We believe the ability for the network to quickly and nimbly handle large data flows at the most cost-effective layer of the network is one of the key value drivers for SDN,” said ESnet Chief Technologist Inder Monga. “Being able to provision and optimize the network across both the router and transport layers from a single console using our OSCARS platform and open standards-based protocols is very attractive to ESnet.”

“The Brocade MLXe Series delivers a high-performance SDN solution ideal for the customers who require scalability and programmability from the network,” said Ananda Rajagopal, Vice President Product Management and Strategy, Brocade. “This multivendor solution showcases the importance of open standards and how they can improve network provisioning across multiple layers of the network for increased operational simplicity.”

“As the steward of the OpenFlow protocol, the Open Networking Foundation (ONF) is excited to see rapid innovation and business value delivered in the carrier market using a standards-based OpenFlow approach,” said Dan Pitt, Executive Director of the Open Networking Foundation. “We are pleased to see ONF member companies propelling the open SDN movement and accelerating industry adoption.”

http://www.infinera.com
http://www.brocade.com/SDN

Read More

Juniper Upgrades its MX Series Edge Routers

Juniper Networks announced line-card enhancements and a new switch fabric for its SDN-ready, MX Series 3D Universal Edge Router portfolio.  The upgrades significantly expand system capacity, subscriber bandwidth and service performance.

The new line cards increase per-slot bandwidth up to 520 Gbps and offer hardware support for as many as 128,000 subscribers, enabling cost-effective subscriber concentration. The new switch fabric module more than doubles the capacity of several key MX 3D platforms, including the widely deployed MX960 and enhanced service cards for the entire MX Series portfolio. These new service cards enable up to three times the per-slot capacity and up to four times the service density of similar modules offered by other vendors.

The enhancements are compatible with tens of thousands of MX Series 3D Universal Edge platforms deployed by service provider, cable and enterprise networks globally. The upgrade path supports SDN capabilities, including those powered by Juniper’s Contrail SDN controller.

Juniper said the programmability of its Junos Trio chipset gives its MX platform unmatched flexibility and hardware investment protection in even the fastest-growing markets while enabling virtualized application support through Juniper's SDN strategy.

The MX Series allows service providers and cable operators to further improve operational efficiently by consolidating appliances and applications directly on the router. Not only is the increased performance of the MX Series 3D important for new customers, but it is also gives every existing MX customer, including those that deployed the first 480G capacity MX960 six years ago, a seamless path to upgrade.

“Leveraging the common, feature-rich Junos operating system and the powerful Trio chipset, the MX portfolio offers customers the ability to create revenue-generating services with complete operational consistency. The MX 3D family serves the entire spectrum of needs from the smallest access aggregation needs with the MX104, up through the very high-end aggregation and core applications with the MX2020, which is the highest capacity single chassis router in the industry at 10.4Tbps. This gives network operators a single family of best-in-class products that can be leveraged in mobile and fixed businesses, and for residential and business applications,” stated Paul Obsitnik, vice president, product marketing, Platform Systems Division, Juniper Networks.

http://www.juniper.net

• In October 2012, Juniper Networks  introduced its scaled-up MX2020 and MX2010 3D Universal Edge Routers, packing up to 80 Tbps of routing capability in a single chassis and adding new virtualization capabilities for multivendor applications.At the high-end, a single MX2020 3D Universal Edge router offers 80 Tbps of capacity using Juniper's existing Junos Trio Chipset.  Existing line-cards based on the Trio chipset can be added to the new chassis. The MX2010 3D router supports 40 Tbps of capacity. Juniper also introduced a JunosV App Engine for the MX Series.  This hypervisor allows Service Providers to run various applications simultaneously in isolation, such as load balancing, security services, online gaming and advertising. The MX Virtual Chassis can virtualize broadband or business edge services over multiple MX 3D routers allowing service providers to operate the systems as a single chassis and quickly add new subscribers.

Read More

Coho Data Comes Out of Stealth with Cloud-Scale SDN Storage

Coho Data, a start-up based in Sunnyvale, California with R&D in Vancouver (Canada) and Cambridge (UK), emerged from stealth mode to introduce a new storage architecture that combines flash arrays for performance, COTS server hardware for scale-out sizing, and SDN intelligence.

coho Data, which was founded by key contributors to the Xen ecosystem, is taking a software-defined approach that combines commodity hardware with software that’s tuned for the latest PCIe flash technologies. The company said its team leveraged the experience of building the virtualization layer that powers Amazon’s EC2.  The goal is an on-premise storage solution for data centers that delivers twice the price/performance of all-flash arrays. OpenFlow is used to embed storage intelligence in the network. A 2U Coho DataStream appliance provides Up to 180K IOPS and can be scaled linearly.

The Coho DataStream product line is currently installed in large enterprise private cloud environments and will be general available later this year.

Coho Data received initial funding from Andreessen Horowitz.

"As we saw with Nicira and networking, Coho Data is shaking up how storage has traditionally been delivered; instead of focusing on the storage hardware, it's all about using software to extract the best performance from commodity hardware. I'm thrilled to work with the founding team, comprised of veteran product visionaries and architects, to usher in storage for the cloud generation," stated Peter Levine, Andreessen Horowitz General Partner.

“Even as new advancements in flash memory have come on the scene, the storage industry has remained stagnant, relying on a 30 year old architecture to deliver performance and accessibility,” said Ramana Jonnala, CEO, Coho Data. “By redesigning the storage stack itself, we have taken the best ideas from public cloud-based architectures and improved them for demanding on-premise datacenters. DataStream uses sophisticated software to take advantage of flash in such a way that it can be used for all applications, not just the top tier ones. Whether data is in the public or private cloud, it should be on a storage architecture that can meet the scalability and performance today’s cloud generation companies need, at the pricing they demand.”

“We deliver services to a range of internal customers and require a storage solution that can provide cloud-level scale to help us manage the 12PBs of data and 4K VMs we currently have,” said Mario Angers, Systems Manager for University of British Columbia Central IT. “We’ve looked at lots of flash solutions, but they deliver performance that doesn’t scale. Coho DataStream is the first flash-tuned architecture we’ve seen that addresses scale and performance for our large VMware environment.”

http://www.cohodata.com

Read More

Procera's PacketLogic 8960 Policy Enforcement Platform Packs 7 0Gbps Throughput

Procera Networks introduced its PacketLogic 8960, a 2RU flexible Intelligent Policy Enforcement (IPE) appliance for network deployments supporting high-density subscriber scenarios.The PacketLogic 8960 supports up to 70 Gbps of network traffic throughput accessible through 6 modular I/O slots supporting up to 12 x 10GE or 24 x 1GE channels. It supports up to three million active subscribers and 20 million active flows, with the ability to enforce policy on all concurrent flows in the system.

Procera said its new PL8960 is designed for both larger fixed broadband service provider deployments where traffic is aggregated and analyzed at the edge of the network, requiring high I/O density in a small form factor, and in the core network of more geographically compact service providers. The PL8960 is also ideally suited to LTE network deployments where mobile operators need dispersed, small footprint configurations that quickly and economically scale to millions of users as LTE coverage expands to support subscriber expansion.

"With the introduction of the PL8960, Procera Networks once again sets a new industry benchmark for implementing IPE capabilities that brings chassis-level scale, performance, and reliability into a space and energy efficient appliance form factor," said Cam Cullen, vice president of Global Marketing for Procera Networks. "Our new PL8960 is a natural evolution of our highly regarded PL8920 platform that has been widely deployed by both fixed broadband service providers and mobile operators worldwide."

http://www.proceranetworks.com

Read More

Russia's TransTeleCom Launches 100G with Cisco

TransTeleCom (TTK), one of Russia’s largest telecom operators, announced the commercial launch of its 100G Ultra Long Haul (ULH) DWDM network in the St. Petersburg-Moscow and Moscow-Chelyabinsk-Yekaterinburg segments.

The coherent optical technology increases the backbone network capacity from 40 to 100G per optical carrier. The network uses Cisco's optical technology.

Launch of the segments with a total length of over 3000 km completes the first stage of the company’s project to create a national 100G ULH DWDM network. The operational segment amounts to about 20 percent of the whole TTK 100G network under construction.

“TTK’s backbone network is the most advanced in Russia and is a crucial part of the country’s essential telecom infrastructure; its constant development and extension is of utmost importance not only for the company, but also for the whole communications market. We are pleased to cooperate with TTK in this project, and I’m sure that it will be long-term and productive,” stated Bill Gartner, vice-president, High-end Routing and Optical Group, Cisco.

TTK's main shareholder is Russian railways.

http://www.cisco.com

Read More

Alcatel-Lucent Workers Protest in Paris

More than 1,500 employees of Alcatel-Lucent marched in a public protest in Paris over the company's recently announced plans to reduce its workforce by 10,000 by the end of 2015.  About 10% of the cuts are to occur in France, which the protesters said is too much.  The company is to begin negotiation with work councils soon.

• Earlier this month. Alcatel-Lucent said the coming he job cuts would include a reduction of 4,100 positions in Europe, Middle East and Africa, 3,800 in Asia Pacific and 2,100 in Americas. By the end of 2015, Alcatel-Lucent will halve the number of its business hubs globally.
  
• In France, Alcatel-Lucent plans to cut approximately 900 positions in 2014 primarily in support, administrative and sales functions (via a legally-compliant program know in France as Plan de Sauvegarde pour l’Emploi), as well as recruiting 200 engineers and technicians with new technical competencies. By the end of 2015, the company transformation program could also result in internal mobility, transfers to partners and redeployments for approximately 900 employees whose jobs will be retained, inside or outside the company.
  
• Alcatel-Lucent said its R&D activities in France will focus on 4G and IP platforms, in particular with the creation of a new small cells competency center, an area of particular interest for the company. In terms of research, France will keep its focus on  optics and strengthen it in mathematics, at the heart of next-generation network software.
  
• In June 2013, Alcatel-Lucent outlined a "Shift Plan" aimed at transforming the company from a generalist supplier of networking solution into a specialist provider of IP Networking and Ultra-Broadband Access.  The company is targeting Euro 1 billion in reduced sales, general and administrative (SG&A) expenses over the next three years as it makes a decisive in its industrial focus toward high-value equipment and services. Current business segments outside of this new area of focus include Terrestrial Optics, Submarine, Wireless Transmission, Network Applications, Integration Services for Strategic Industries, Consulting Businesses and OSS/BSS.

Read More

IEEE Comments on Trust in Standards Setting Process

On the observation of World Standards Day, the IEEE now is an opportune moment to denote the importance of

The IEEE published a commentary discussing current erosion of trust in the technical foundations of the Internet. Reflecting on the open standardization processes and the principles of OpenStand, the modern standards paradigm that has driven technological and economic advances in recent decades, the IEEE said the standardization processes must be universally open and fully transparent, especially in the domain of cyber security.

"The principles of direct participation, independent of the passport or where one is domiciled, of due process, of broad consensus, of balance, of transparency and of universal openness from the very beginning of the standardization process are practiced by thousands of collaborative groups around the world, building a kind of global basis democracy in virtually all fields of technology."

The commentary highlights six principles:

1. Direct participation means there are no intermediaries between a good idea and the peer group that decides the start of a new standardization project. Everybody, from anywhere can submit a proposal without permission by a government or a company or elsewhere.

2. Due process provides for a level playing field for all participants to understand the rules of engagement; the opportunity for equal participation; and provides a means for the definition of a standard based upon technical merit.

3. Broad consensus ensures that decisions are made by either a majority or supermajority of participants, so no single person or entity wields undue power in the definition of a standard.

4. Balance provides opportunity for a multitude of stakeholders to participate in the definition of a standard, resulting in global interoperability, scalability, stability, and resiliency. This, in turn, enables global competition and outputs that serve as building blocks for further innovation, thus contributing to the creation of benefit for humanity.

5. Transparency of the process and proceedings ensure that the policies and procedures under which the standard is defined are available so that participants understand the rules of engagement and so that appropriate audit trails are available for inspection upon emergence of an output.

6. Universal openness from the very beginning ensures that the appropriate notification to a global audience is provided for attendance at meetings and engagement in the definition of a standard. Combined with transparency, this creates the basis for a robust and trustworthy standardization process in all fields of technology, including for cybersecurity and privacy.

http://www.open-stand.org
http://www.ieee.org

Read More

CipherCloud Delivers Breakthrough Searchable Strong Encryption

CipherCloud announced new techniques to improve the searchability, usability and security of cloud data protected with AES 256-bit encryption.

CipherCloud's Searchable Strong Encryption (SSE) solution provides organizations with the ability to strongly encrypt their sensitive information in real time before it's sent to the cloud and exclusively retain their encryption keys. The solution also enables organization to comply with government regulations and industry mandates including – GLBA, PCI, HIPAA and HITECH, the EU Data Protection Act, UK ICO guidance, the Australian Privacy Amendment Act and US State Privacy laws.

CipherCloud's SSE technology leverages its gateway architecture to provide secure local index and search operations while sending the strongly encrypted data to the cloud and protecting it from all external threats. This solution enables natural language, wild cards and Boolean searches of AES 256-bit encrypted data. Other available techniques on the market lack searchability or require complex deployments of local databases or rely only on partial data encryption.

"Business users demand security to be transparent," said Pravin Kothari, founder and CEO of CipherCloud. "Applying strong encryption, AES 256-bit, to data while keeping it entirely searchable has been a long-standing challenge. CipherCloud has now raised the bar by enabling the full usability of encrypted data in the cloud, without compromising security or performance."

CipherCloud delivers full protection of an organization's structured and unstructured information in popular cloud applications including – salesforce.com, Box, Microsoft Office 365, Google Gmail, Amazon Web Services and many more. Additionally, CipherCloud for Any App and CipherCloud for Databases enable organizations to extend data protection to hundreds of third-party cloud and private cloud applications and databases.

http://www.ciphercloud.com

Read More