Marvell Blog

Featuring technology ideas and solutions worth sharing

Marvell

Latest Articles

June 29th, 2018

Marvell Helps to Bring HD Video Capabilities to New Entry-Level Drone

By Sree Durbha, Head of Smart-Connected Business, Marvell

The consumer drone market has expanded greatly over the last few years, with almost 3 million units shipped during 2017. This upward trend is likely to continue. Analyst firm Statista forecasts that the commercial drone business will be worth $6.4 billion annually by 2020, while Global Market Insights has predicted that the worldwide drone market will grow to $17 billion (with the consumer category accounting for $9 billion of that). As new products are continually being introduced into what is already an acutely overcrowded marketplace, a differentiated offering is therefore critical to a successful product.

One of the newest and most exciting entrants into this crowded drone market, Tello, features functionality that sets it apart from rival offerings. Tello is manufactured by Shenzhen-based start-up Ryze Tech, a subsidiary of well-known brand DJI, which is the world’s largest producer of drones and unmanned aerial vehicles (UAVs). With a 13 minute runtime, plus a flight distance of up to 100 meters, this is an extremely maneuverable and compact quadcopter drone. It weighs just 80 grams and can fit into the palm of a typical teenager’s hand (with dimensions of 98 x 92.5 x 41 millimeters). The two main goals of the Tello are fun and education. To that end, a smartphone App-based control provides a fun user interface for everyone, including young people, to play with. The educational goal is met through an easy to program visual layout that allows users to write their own code using the comprehensive software development kit (SDK) included in the package. What really distinguishes Tello from other drones, however, is the breadth of its imaging capabilities – and this is where engaging with Marvell has proven pivotal.

Tello’s original drone design requirement called for livestreaming 720p MP4 format video, using its 5 Megapixel image sensor, back to the user’s smartphone or tablet even while traveling at its maximum speed of 8 meters/second. This called for interoperability testing with a broad array of smartphone and tablet models. Due to its small size, conserving battery life would be a key requirement, which meant ultra-low power consumption by Wi-Fi®. Underlying all of this was the singular requirement for a strong wireless connection to be maintained at all times. Finally, as is always the case, Wi-Fi would need to fit in the low bill of materials for the product.

Initial discussions between technical teams at Ryze and Marvell revealed a perfect match between the features offered on the Marvell® 1×1 802.11n single-band Wi-Fi system-on-chip (SoC) and the Wi-Fi requirements for the Tello drone project. This chip was already widely adopted in the market and established itself as a proven solution for various customer applications, including video transmission in IP cameras, mobile routers, IoT gateways etc. Ryze chose this chipset, banking on its reliability while transmitting high-definition video over the air, exceptional RF performance over range while offering ultra-low power operation, all at a competitive price point.

Marvell’s Wi-Fi SoC is a highly integrated, single-band (2.4GHz) IC that delivers IEEE® 802.11b/g/n operation in a single spatial stream (1 SS) configuration. It incorporates a power amplifier (PA), a low noise amplifier (LNA) and a transmit/receive switch. Quality of Service (QoS) is guaranteed through the 802.11e standard implementation. The Wi-Fi SoC’s compliance with the 802.11i security protocol, plus built-in wired equivalent privacy (WEP) algorithms, enable 128-bit encryption of transmitted data, thereby protecting the data from being intercepted by third parties. All of these hardware features are supported by Marvell’s robust Wi-Fi software, which includes a small footprint and full featured Wi-Fi firmware tied in with the hardware level features. Specific features such as infrastructure mode operation were developed to enable the functionality desired by Ryze for the Tello.

Marvell’s industry-leading Wi-Fi technology has enabled an exciting new user experience in the Tello, at a level of sophistication that previously would only have been seen in expensive, professional-grade equipment. In order to bring this professional quality experience to an entry-level drone model meant that significant power, performance and cost barriers were overcome. As we enter the 802.11ax era of Wi-Fi industry transition, expect Marvell to launch first-to-market, ever more envelope-pushing, technological advances such as uplink OFDMA.

 

 

June 7th, 2018

Versatile New Ethernet Switch Simultaneously Addresses Multiple Industry Sectors

By Ran Gu, Marketing Director of Switching Product Line, Marvell

Due to ongoing technological progression and underlying market dynamics, Gigabit Ethernet (GbE) technology with 10 Gigabit uplink speeds is starting to proliferate into the networking infrastructure across a multitude of different applications where elevated levels of connectivity are needed: SMB switch hardware, industrial switching hardware, SOHO routers, enterprise gateways and uCPEs, to name a few. The new Marvell® Link Street™ 88E6393X, which has a broad array of functionality, scalability and cost-effectiveness, provides a compelling switch IC solution with the scope to serve multiple industry sectors.

The 88E6393X switch IC incorporates both 1000BASE-T PHY and 10 Gbps fiber port capabilities, while requiring only 60% of the power budget necessitated by competing solutions. Despite its compact package, this new switch IC offers 8 triple speed (10/100/1000) Ethernet ports, plus 3 XFI/SFI ports, and has a built-in 200 MHz microprocessor. Its SFI support means that the switch can connect to a fiber module without the need to include an external PHY – thereby saving space and bill-of-materials (BoM) costs, as well as simplifying the design. It complies with the IEEE 802.1BR port extension standard and can also play a pivotal role in lowering the management overhead and keeping operational expenditures (OPEX) in check. In addition, it includes L3 routing support for IP forwarding purposes.

Adherence to the latest time sensitive networking (TSN) protocols (such as 802.1AS, 802.1Qat, 802.1Qav and 802.1Qbv) enables delivery of the low latency operation mandated by industrial networks. The 256 entry ternary content-addressable memory (TCAM) allows for real-time, deep packet inspection (DPI) and policing of the data content being transported over the network (with access control and policy control lists being referenced). The denial of service (DoS) prevention mechanism is able to detect illegal packets and mitigate the security threat of DoS attacks.

The 88E6393X device, working in conjunction with a high performance ARMADA® network processing system-on-chip (SoC), can offload some of the packet processing activities so that the CPU’s bandwidth can be better focused on higher level activities. Data integrity is upheld, thanks to the quality of service (QoS) support across 8 traffic classes. In addition, the switch IC presents a scalable solution. The 10 Gbps interfaces provide non-blocking uplink to make it possible to cascade several units together, thus creating higher port count switches (16, 24, etc.).

This new product release features a combination of small footprint, lower power consumption, extensive security and inherent flexibility to bring a highly effective switch IC solution for the SMB, enterprise, industrial and uCPE space.

 

May 31st, 2018

Why is 802.11ax a “must have” for the connected car?

By Avinash Ghirnikar, Director of Technical Marketing of Connectivity Business Group, Marvell

Imagine motoring along through busy, urban traffic in your new connected car that is learning, getting smarter, safer and more reliable as it is driving. Such a car is constantly gathering and generating all kinds of data that is intermittently and opportunistically being uploaded to the cloud. As more cars on the road feature advanced wireless connectivity, this exciting future will become commonplace. However, each car will need to share the network with potentially hundreds of other cars that might be in its vicinity.

While such a use case could potentially rely on LTE/5G cellular technology, the costs associated with employing such a “licensed pipe” would be prohibitively expensive. In such situations, the new Wi-Fi® standard 802.11ax, also known as high efficiency wireless (HEW), will be a life saver for the automotive industry. The zettabytes of data that cars equipped with a slew of sensors will create in the years to come will all need to be uploaded to the cloud and data centers, enabling next-generation machine learning in order to make driving increasingly safe and predictable in the future. Uploading this data will, of course, need to be done both securely and reliably.

The car – as an 802.11ax station (STA) – will also be to able upload data to an 802.11ax access point (AP) in the most challenging of wireless environments while sharing the network with other clients. The 802.11ax system will be able to do this via technologies like MU-MIMO and OFDMA (allowing for spatial, frequency and time reuse) which are new innovations that are part of this emerging standard. Today, STAs compete rather than effectively share the network and have to deal with the dreaded “circle of death”’ awaiting connectivity. This is because today’s wireless standard can often be in an all-or-nothing binary mode of operation due to constant competition. When coupled with other upcoming standards like 802.11ai, specifically fast initial link setup (FILS), this vision of cars uploading data to the cloud over Wi-Fi becomes a true reality, even in environments where the car is moving and likely hopping from one AP to another.

While this “under the hood” upload use case is greatly enhanced by the 802.11ax standard from an infrastructure perspective, download of software and firmware into connected cars can also be transformed by this same standard. It is well known that the number of processors and electronic control units (ECUs) in car models is expected to increase dramatically. This, in turn, implies that the software/firmware content in these cars will likewise grow at exponential rates. Periodic firmware over-the-air (FOTA) updates will be required and, therefore, having a reliable and robust mechanism to support this will be vital for automobile manufacturers – potentially saving them millions of dollars in relation to servicing costs, etc.  Such is the pace of innovation and technological change these days that this can sometimes happen almost immediately after cars come off the assembly line.

Take the example of a parking lot outside an auto plant containing hundreds of brand new cars requiring some of their software to be updated.  Here, too, 802.11ax can come to the rescue by making a mass update more efficient and reliable. This advantage will then carry forward for the rest of the lifespan of each car, since it can never be predicted what sort of wireless connectivity environment these cars will encounter. These could be challenging environments like garages, driveways, and maybe even parking decks. The modulation enhancements that 802.11ax delivers, coupled with MU-MIMO and OFDMA features, will ensure that the most efficient and reliable Wi-Fi pipe is always available for such a critical function. Given that a car can easily be on the road for close to a decade, having this functionality built in from day one would be a tremendous advantage and could enable significant cost savings. Again, accompanying technologies like Wake on Bluetooth® Low Energy and Bluetooth Low Energy Long Range will also play a pivotal role in ensuring this use case is realized from an overall end-to-end system standpoint.

These two infrastructure type use cases are likely to be tremendous value-adds for the connected car and can justify the presence of 802.11ax, especially from an automobile manufacturers’ point of view. Even consumers are likely to see significant benefits in their vehicle dashboards where the mobile APs in their infotainment systems will be able to seamlessly connect to their latest smartphone handsets (which will themselves be 802.11ax capable within the 2019 timeframe). Use cases like Wireless Apple CarPlay®, Wireless Android Auto™ Projection, rear seat entertainment, wireless cameras, etc. will all be a breeze given the additional 30-40% throughput enhancement in 802.11ax (and the backward compatibility this standard has with previous Wi-Fi standards for such use cases to cooperatively coexist).  Just as in homes, the number of Wi-Fi endpoints in cars is also proliferating. The 802.11ax standard is the only well-designed path for an increasing number of endpoints and yet provides the best user experience.

The 802.11ax as Release 1 (aka Wave 1) is well on its way to a concrete launch by the Wi-Fi Alliance in the second half of 2019. Products are already being sampled by silicon vendors – both on the AP and STA/mobile AP side – and interoperability testing is well underway. For all wireless system designers at OEMs and their Tier 1 suppliers, the 802.11ax Wi-Fi standard should be a goal, and especially for any product launch set for 2020 and beyond.  The time has come to begin future proofing for the impending arrival of 802.11ax infrastructure. The days of the wireless technology in your smartphone/home/enterprise and in your car belonging to different generations are long gone. Consumers demand that their cars now be an extension of their home/work environments and that all of these living spaces function as one. The 802.11ax is destined to be one of the key pillars of technology to make such a vision a reality.

Marvell has been a pioneer in designing Wi-Fi/Bluetooth combo devices for the automotive market since the debut of such devices in cars in 2011. With actual development beginning almost a decade ago, Marvell’s automotive wireless portfolio has been honed to address key use cases over five generations of products, through working closely with OEMs, Tier 1s and Tier 2s. All the technologies needed to achieve the various use cases described above have been incorporated into Marvell’s fifth generation device. Coupled with Marvell’s offering for enterprise class, high-performance APs, Marvell remains committed to providing the automobile industry and car buyers with the best wireless connectivity experience — encompassing use cases inside and outside of the car today, and well into the future.

 

May 2nd, 2018

Cavium FastLinQ Ethernet Adapters Available for HPE Cloudline Servers

By Todd Owens, Technical Marketing Manager, Marvell

Are you considering deploying HPE Cloudline servers in your hyper-scale environment? If you are, be aware that HPE now offers select Cavium FastLinQ® 10GbE and 10/25GbE Adapters as options for HPE Cloudline CL2100, CL2200 and CL3150 Gen 10 Servers. The adapters supported on the HPE Cloudline servers are shown in table 1 below.

Table 1: Cavium FastLinQ 10GbE and 10/25GbE Adapters for HPE Cloudline Servers

As today’s hyper-scale environments grow, the Ethernet I/O needs go well beyond providing basic L2 NIC connectivity. Faster processors, increase in scale, high performance NVMe and SSD storage and the need for better performance and lower latency have started to shift some of the performance bottlenecks from servers and storage to the network itself. That means architects of these environments need to rethink connectivity options.

While HPE already have some good I/O offerings for Cloudline from other vendors, having Cavium FastLinQ adapters in the portfolio increases the I/O features and capabilities available. Advanced features like Universal RDMA, SmartAN™, DPDK, NPAR and SR-IOV from Cavium, allow architects to design more flexible and scalable hyper-scale environments.

Cavium’s advanced feature set provides offload technologies that shift the burden of managing the I/O from the O/S and CPU to the adapter itself. Some of the benefits of offloading I/O tasks include:

  • Lower CPU utilization to free up resources for applications or more VM scalability
  • Accelerate processing of small-packet I/O with DPDK
  • Save time by automating adapter connectivity between 10GbE and 25GbE
  • Reduced latency through direct memory access for I/O transactions to increase performance
  • Network isolation and QoS at the VM level to improve VM application performance
  • Reduce TCO with heterogeneous management

To deliver these benefits, customers can take advantage of some or all the advanced features in the Cavium FastLinQ Ethernet adapters for HPE Cloudline. Here’s a list of some of the technologies available in these adapters.

* Source; Demartek findings 
Table 2: Advanced Features in Cavium FastLinQ Adapters for HPE Cloudline

Network Partitioning (NPAR) virtualizes the physical port into eight virtual functions on the PCIe bus. This makes a dual port adapter appear to the host O/S as if it were eight individual NICs. Furthermore, the bandwidth of each virtual function can be fine-tuned in increments of 500Mbps, providing full Quality of Service on each connection. SR-IOV is an additional virtualization offload these adapters support that moves management of VM to VM traffic from the host hypervisor to the adapter. This frees up CPU resources and reduces VM to VM latency.

Remote Direct Memory Access (RDMA) is an offload that routes I/O traffic directly from the adapter to the host memory. This bypasses the O/S kernel and can improve performance by reducing latency. The Cavium adapters support what is called Universal RDMA, which is the ability to support both RoCEv2 and iWARP protocols concurrently. This provides network administrators more flexibility and choice for low latency solutions built with HPE Cloudline servers.

SmartAN is a Cavium technology available on the 10/25GbE adapters that addresses issues related to bandwidth matching and the need for Forward Error Correction (FEC) when switching between 10Gbe and 25GbE connections. For 25GbE connections, either Reed Solomon FEC (RS-FEC) or Fire Code FEC (FC-FEC) is required to correct bit errors that occur at higher bandwidths. For the details behind SmartAN technology you can refer to the Cavium technology brief here.

Support for Data Plane Developer Kit (DPDK) offloads accelerate the processing of small packet I/O transmissions. This is especially important for applications in the Telco NFV and high-frequency trading environments.

For simplified management, Cavium provides a suite of utilities that allow for configuration and monitoring of the adapters that work across all the popular O/S environments including Microsoft Windows Server, VMware and Linux. Cavium’s unified management suite includes QCC GUI, CLI and v-Center plugins, as well as PowerShell Cmdlets for scripting configuration commands across multiple servers. Cavum’s unified management utilities can be downloaded from www.cavium.com .

Gen10 servers. Each of the Cavium adapters shown in table 1 support all of the capabilities noted above and are available in standup PCIe or OCP 2.0 form factors for use in the HPE Cloudline Gen10 Servers. One question you may have is how do these adapters compare to other offerings for Cloudline and those offered in HPE ProLiant servers? For that, we can look at the comparison chart here in table 3.

 

Table 3: Comparison of I/O Features by Ethernet Supplier

Given that Cloudline is targeted for hyper-scale service provider customers with large and complex networks, the Cavium FastLinQ Ethernet adapters for HPE Cloudline offer administrators much more capability and flexibility than other I/O offerings. If you are considering HPE Cloudline servers, then you should also consider Cavium FastLinQ as your I/O of choice.

April 30th, 2018

ARMADA 3720 SoC Enables Ground-Breaking Modular Router from CZ.NIC

By Maen Suleiman, Senior Software Product Line Manager, Marvell

Marvell ARMADA® embedded processors are part of another exciting networking solution for a crowdfunding project and are helping “power” the global open hardware and software engineering community as innovative new products are developed. CZ.NIC, an open source networking research team based in the Czech Republic, just placed its Turris MOX modular networking appliance on the Indiegogo® platform and has already obtained over $110,000 in financial backing.

MOX has a highly flexible modular arrangement. Central to this is a network processing module featuring a Marvell® ARMADA 3720 network processing system-on-chip (SoC). This powerful yet energy efficient 64-bit device includes dual Cortex®-A53 ARM® processor cores and an extensive array of high speed IOs (PCIe 2.0, 2.5 GbE, USB 3.0, etc.).

Figure 1: The MOX Solution from CZ.NIC

The MOX concept is simple to understand. Rather than having to procure a router with excessive features and resources that all add to the cost but actually prove to be superfluous, users can just buy a single MOX that can subsequently be extended into whatever form of network appliance a user needs. Attachment of additional modules means that specific functionality can be provided to meet exact user expectations. There is an Ethernet module that adds 4 GbE ports, a fiber module that adds fiber optic SFP connectivity, and an extension module that adds a mini PCIe connection. At a later stage, if requirements change, it is possible for that same MOX to be repurposed into a completely different appliance by adding appropriate modules.


                        Figure 2: The MOX Add-On Modules – Base, Extension, Ethernet and SFP

The MOX units run on Turris OS, an open source operating system built on top of the extremely popular OpenWrt® embedded Linux® distribution (as supported by Marvell’s ARMADA processors). This gives the appliance a great deal of flexibility, allowing it to execute a wide variety of different networking functions that enable it to operate as an email server, web server, firewall, etc. Additional MOX modules are already under development and will be available soon.

This project follows on CZ.NIC’s previous crowdfunding campaign using Marvell’s ARMADA SoC processing capabilities for the Turris Omnia high performance open source router – which gained huge public interest and raised 9x its original investment target. Turris MOX underlines the validity of the open source software ecosystem that has been built up around the ARMADA SoC to help customers bring their ideas to life.

Click here to learn more on this truly unique Indiegogo campaign.

 

April 5th, 2018

VMware vSAN ReadyNode Recipes Can Use Substitutions

By Todd Owens, Technical Marketing Manager, Marvell

VMware vSAN ReadyNode Recipes Can Use Substitutions

When you are baking a cake, at times you substitute in different ingredients to make the result better. The same can be done with VMware vSAN ReadyNode configurations or recipes. Some changes to the documented configurations can make the end solution much more flexible and scalable.

VMware allows certain elements within a vSAN ReadyNode bill of materials (BOM) to be substituted. In this VMware BLOG, the author outlines that server elements in the bom can change including:

  • CPU
  • Memory
  • Caching Tier
  • Capacity Tier
  • NIC
  • Boot Device

However, changes can only be made with devices that are certified as supported by VMware. The list of certified I/O devices can be found on VMware vSAN Compatibility Guide and the full portfolio of NICs, FlexFabric Adapters and Converged Network Adapters form HPE and Cavium are supported.

If we zero in on the HPE recipes for vSAN ReadyNode configurations, here are the substitutions you can make for I/O adapters.

Ok, so we know what substitutions we can make in these vSAN storage solutions. What are the benefits to the customer for making this change?

There are several benefits to the HPE/Cavium technology compared to the other adapter offerings.

  • HPE 520/620 Series adapters support Universal RDMA – the ability to support both RoCE and IWARP RDMA protocols with the same adapter.
    • Why Does This Matter? Universal RDMA offers flexibility in choice when low-latency is a requirement. RoCE works great if customers have already deployed using lossless Ethernet infrastructure. iWARP is a great choice for greenfield environments as it works on existing networks, doesn’t require complexity of lossless Ethernet and thus scales infinitely better.
  • Concurrent Network Partitioning (NPAR) and SR-IOV
    • NPAR (Network Partitioning) allows for virtualization of the physical adapter port. SR-IOV Offloadmove management of the VM network from the Hypervisor (CPU) to the Adapter. With HPE/Cavium adapters, these two technologies can work together to optimize the connectivity for virtual server environments and offload the Hypervisor (and thus CPU) from managing VM traffic, while providing full Quality of Service at the same time.
  • Storage Offload
    • Ability to reduce CPU utilization by offering iSCSI or FCoE Storage offload on the adapter itself. The freed-up CPU resources can then be used for other, more critical tasks and applications. This also reduces the need for dedicated storage adapters, connectivity hardware and switches, lowering overall TCO for storage connectivity.
  • Offloads in general – In addition to RDMA, Storage and SR-IOV Offloads mentioned above, HPE/Cavium Ethernet adapters also support TCP/IP Stateless Offloads and DPDK small packet acceleration offloads as well. Each of these offloads moves work from the CPU to the adapter, reducing the CPU utilization associated with I/O activity. As mentioned in my previous blog, because these offloads bypass tasks in the O/S Kernel, they also mitigate any performance issues associated with Spectre/Meltdown vulnerability fixes on X86 systems.
  • Adapter Management integration with vCenter – All HPE/Cavium Ethernet adapters are managed by Cavium’s QCC utility which can be fully integrated into VMware v-Center. This provides a much simpler approach to I/O management in vSAN configurations.

In summary, if you are looking to deploy vSAN ReadyNode, you might want to fit in a substitution or two on the I/O front to take advantage of all the intelligent capabilities available in Ethernet I/O adapters from HPE/Cavium. Sure, the standard ingredients work, but the right substitution will make things more flexible, scalable and deliver an overall better experience for your client.

April 2nd, 2018

Understanding Today’s Network Telemetry Requirements

By Tal Mizrahi, Feature Definition Architect, Marvell

There have, in recent years, been fundamental changes to the way in which networks are implemented, as data demands have necessitated a wider breadth of functionality and elevated degrees of operational performance. Accompanying all this is a greater need for accurate measurement of such performance benchmarks in real time, plus in-depth analysis in order to identify and subsequently resolve any underlying issues before they escalate.

The rapidly accelerating speeds and rising levels of complexity that are being exhibited by today’s data networks mean that monitoring activities of this kind are becoming increasingly difficult to execute. Consequently more sophisticated and inherently flexible telemetry mechanisms are now being mandated, particularly for data center and enterprise networks.

A broad spectrum of different options are available when looking to extract telemetry material, whether that be passive monitoring, active measurement, or a hybrid approach. An increasingly common practice is the piggy-backing of telemetry information onto the data packets that are passing through the network. This tactic is being utilized within both in-situ OAM (IOAM) and in-band network telemetry (INT), as well as in an alternate marking performance measurement (AM-PM) context.

At Marvell, our approach is to provide a diverse and versatile toolset through which a wide variety of telemetry approaches can be implemented, rather than being confined to a specific measurement protocol. To learn more about this subject, including longstanding passive and active measurement protocols, and the latest hybrid-based telemetry methodologies, please view the video below and download our white paper.

WHITE PAPER, Network Telemetry Solutions for Data Center and Enterprise Networks