Marvell Blog

Featuring technology ideas and solutions worth sharing.

Archive for the ‘Enterprise Services’ Category

January 9th, 2017

Ethernet Darwinism: The survival of the fittest (or the fastest)

By Michael Zimmerman

The most notable metric of Ethernet technology is the raw speed of communications. Ethernet has taken off in a meaningful way with 10BASE-T, which was used ubiquitously across many segments. With the introduction of 100BASE-T, the massive 10BASE-T installed base was replaced, showing a clear Darwinism effect of the fittest (fastest) displacing the prior and older generation. However, when 1000BASE-T (GbE – Gigabit Ethernet) was introduced, contrary to industry experts’ predictions, it did not fully displace 100BASE-T, and the two speeds have co-existed for a long time (more than 10 years). In fact, 100BASE-T is still being deployed in many applications. The introduction — and slow ramp — of 10GBASE-T has not impacted the growth of GbE, and it is only recently that GbE ports began consistently growing year over year. This trend signaled a new evolution paradigm of Ethernet: the new doesn’t replace the old, and the co-existence of multi variants is the general rule. The introduction of 40GbE and 25GbE augmented the wide diversity of Ethernet speeds, and although 25GbE was rumored to displace 40GbE, it is expected that 40GbE ports will still be deployed over the next 10 years1.

Ethernet-Diversity

Hence, a new market reality evolved: there is less of a cannibalizing effect (i.e. newer speed cannibalizing the old), and more co-existence of multiple variants. This new diversity will require a set of solutions which allow effective support for multiple speed interconnect. Two critical capabilities will be needed:

  1. Ability to economically scale-down to a few ports2
  2. Support of multiple Ethernet speeds

Marvell launched a new set of Ethernet interconnect solutions that meet this evolution pattern. The first products in the family are the Prestera® 98DX83xx 320G interconnect switch, and the Alaska® 88X5113 25G/40G Gearbox PHY. The 98DX83xx switch fans-out up to 32-ports of 10GbE or 8-ports of 40GbE, in economical 24x20mm package, with power of less than 0.5Watt/10G port.

Interconnect1

The 88X5113 Gearbox converts a single port of 40GbE to 25GbE. The combination of the two devices creates unique connectivity configurations for a myriad of Ethernet speeds, and most importantly enables scale down to a few ports. While data center- scale 25GbE switches have been widely available for 64-ports, 128-ports (and beyond), a new underserved market segment evolved for a lower port count of 25GbE and 40GbE. Marvell has addressed this space with the new interconnect solution, allowing customers to configure any number of ports to different speeds, while keeping the power envelope to sub-20Watt, and a fraction of the  hardware/thermal footprint of comparable data center solutions. The optimal solution to serve low port count connectivity of 10GbE, 25GbE, and 40GbE is now well addressed by Marvell. Samples and development boards with SDK are ready, with the option of a complete package of application software.

Interconnect2

  1. In the mega data center market, there is cannibalization effect of former Ethernet speeds, and mass migration to higher speeds. However, in the broader market which includes private data centers, enterprise, carriers, multiple Ethernet speeds co-exist in many use cases.
  2. Ethernet switches with high port density of 10GbE and 25GbE are generally available. However, these solutions do not scale down well to sub-24 ports, where there is pent-up demand for devices as proposed here by Marvell.

June 1st, 2016

Marvell Supports An Enterprise-Grade Network Operating System With The Linux Foundation Project

By Yaniv Kopelman

As organizations continue to invest in data centers to host a variety of applications, more demands are placed on the network infrastructure. The deployment of white boxes is an approach organizations can take to meet their networking needs. White box switches are a “blank” standard hardware that relies on a network operating system, commonly Linux-based. White box switches coupled with Open Network Operating Systems enable organizations to customize the networking features that support their objectives and streamline operations to fit their business.

Marvell is committed to powering the key technologies in the data center and the enterprise network, which is why we are proud to be a contributor to the OpenSwitch Project by The Linux Foundation. Built to run on Linux and open hardware, OpenSwitch is a full-featured network operating system (NOS) aimed at enabling the transition to disaggregated networks. OpenSwitch allows for freedom of innovation while maintaining stability and limiting vulnerability, and has a reliable architecture focused on modularity and availability. The open source OpenSwitch NOS allows developers to build networks that prioritize business-critical workloads and functions, and removes the burdens of interoperability issues and complex licensing structures that are inherent in proprietary systems.

As a provider of switches and PHYs for data center and campus networking markets, Marvell believes that the open source OpenSwitch NOS will help the deployment of white boxes in the data center and campus networks. Developers will be able to build networks that prioritize business-critical workloads and functions, removing the burdens of interoperability issues and complex licensing structures that are inherent in proprietary systems.

Our first contribution is to port the Marvell Switch Software Development Kit (CPSS) to support OpenSwitch. This contribution to an industry standard NOS will enable Marvell devices to be widely used across different markets and boxes.

To learn more, please visit: http://www.openswitch.net/, or read the full press release.

September 29th, 2014

The Next Wave of Flash Storage in the Data Center—Can it Be Client?

By Iri Trashanski

SSD Data StorageWhat happens in an Internet minute?  According to the amalgamation of research by Intel, more than six million Facebook messages are sent, 34.7 million instant messages, 4.1 million Google search inquiries, 138,889 hours of video are watched on YouTube, 23,148 hours of video watched on Netflix and more than $133,000 in Amazon sales. In fact, the total of all digital data created reached four zettabytes (ZB) in 2013. (1ZB = 1,0007 bytes.)

The tens of billions of mobile devices now connected to the Internet will continue to grow as the Internet of Things (IoT) becomes a reality, leading to exponential growth needed for data storage. In fact, according to IDC, data storage is predicted to increase 50-fold, from 0.8 ZB in 2009 to 44ZB by 2020.1 It is also predicted that 32 billion IoT devices will be connected by 2020, 40 percent of the data will be touched by the Cloud, and enterprises will have liability and responsibility for 85 percent of all data.  So how will enterprises keep up?  At Marvell, we believe that in the next four years, the majority of non-client SSD volume will be used in hyperscale public and private cloud deployments.

Hyperscale computing refers to the facilities and provisioning required in distributed computing environments to efficiently scale from a few servers to thousands of servers. Hyperscale favors a stripped-down product design that is extremely cost effective, with a minimal level of investment in hardware and more emphasis on software. In other words, companies will be willing to invest more time to save money versus more money (in hardware, SW and support) to save time. We also believe this trend will continue in data center storage as data centers continue to move toward lower-cost client SSDs to carry the load. This will be achieved with the high-speed PCIe interface.  In fact, with the right high-performance PCIe client-based SSD, we believe more than half the SSDs sold for data center will be PCIe by 2018.

The hyperscale model will be different from the current enterprise.  Hyperscale will focus more on software-oriented growth, huge capacity growth, CapEx and OpEx efficiencies and interactive applications support.  SATA SSDs are successful in data centers because they can be very cost-effective and use client-proven IP.  They are also easy to install in the storage infrastructure. But the SATA interface has been completely saturated, even in client applications.  PCIe SSDs offer 6x the bandwidth, 4x IOPs and 8x improved latency. On top of that, NVM Express (NVMe) has the benefit of very low command and driver overhead. NVMe also enables solutions to connect NAND storage to servers without the need of protocol translations.

Further, hyperscale will require the continuous lowering of the cost of NAND, such as triple-level (TLC) NAND, which requires advanced error correction technologies like low-density parity check (LDPC).  So  squeezing the cost out of SSDs, will take a certain kind of PCIe SSD controller. Marvell’s recent announcement of its next-generation high performance PCIe 3.0 NVMe SSD controller, the Marvell® 88SS1093, advances data center and client solid-state storage with Flash-optimized architecture and performance that will help once again push the price/performance curve in the data center and support the trend toward client SSDs in the data center.

Supporting up to four lanes of PCIe 3.1, the Marvell 88SS1093 SSD controller is suitable for full 4GB/s or 2GB/s end points depending on host system customization.  In order to fully utilize the high-speed PCIe connection, Marvell’s groundbreaking NVMe design facilitates PCIe link data flows by deploying massive hardware automation without compromise. Highly efficient data flow management eliminates legacy host control bottlenecks to finally achieve full flash storage performance.

The Marvell 88SS1093 also integrates Marvell’s third generation NANDEdge™ error-correcting LDPC technology for higher reliability and endurance boost that can support 15nm TLC and 3D NAND, resulting in the enabling of low cost SSDs.

These types of progressions in SSD technology support future trends in the data center that will rely on a distributed, efficient, reliable, scalable and economical architecture.  Hardware will be more commodity-based, using client SSDs and open source software.  So when we look to the next wave of flash storage in the data center, we believe it will be client and that technology is here today.

# # #
1 From EMC-sponsored report, “The Digital Universe of Opportunities: Rich Data and the Increasing Value of the Internet of Things,” with research and analysis by IDC.