By Kristin Hehir, Senior Manager, PR and Marketing, Marvell
Marvell and Memblaze were honored with the “Most Innovative Customer Implementation” award at the Flash Memory Summit (FMS), the industry’s largest conference featuring flash memory and other high-speed memory technologies, last week.Powered by the Marvell® Bravera™ SC5 controller, Memblaze developed the PBlaze 7 7940 GEN5 SSD family, delivering an impressive 2.5 times the performance and 1.5 times the power efficiency compared to conventional PCIe 4.0 SSDs and ~55/9us read/write latency1. This makes the SSD ideal for business-critical applications and high-performance workloads like machine learning and cloud computing. In addition, Memblaze utilized the innovative sustainability features of Marvell’s Bravera SC5 controllers for greater resource efficiency, reduced environmental impact and streamlined development efforts and inventory management.
While Fibre Channel (FC) has been around for a couple of decades now, the Fibre Channel industry continues to develop the technology in ways that keep it in the forefront of the data center for shared storage connectivity. Always a reliable technology, continued innovations in performance, security and manageability have made Fibre Channel I/O the go-to connectivity option for business-critical applications that leverage the most advanced shared storage arrays.
A recent development that highlights the progress and significance of Fibre Channel is Hewlett Packard Enterprise’s (HPE) recent announcement of their latest offering in their Storage as a Service (SaaS) lineup with 32Gb Fibre Channel connectivity. HPE GreenLake for Block Storage MP powered by HPE Alletra Storage MP hardware features a next-generation platform connected to the storage area network (SAN) using either traditional SCSI-based FC or NVMe over FC connectivity. This innovative solution not only provides customers with highly scalable capabilities but also delivers cloud-like management, allowing HPE customers to consume block storage any way they desire – own and manage, outsource management, or consume on demand.HPE GreenLake for Block Storage powered by Alletra Storage MP
At launch, HPE is providing FC connectivity for this storage system to the host servers and supporting both FC-SCSI and native FC-NVMe. HPE plans to provide additional connectivity options in the future, but the fact they prioritized FC connectivity speaks volumes of the customer demand for mature, reliable, and low latency FC technology.
By Amir Bar-Niv, VP of Marketing, Automotive Business Unit, Marvell and John Heinlein, Chief Marketing Officer, Sonatus and Simon Edelhaus, VP SW, Automotive Business Unit, Marvell
The software-defined vehicle (SDV) is one of the newest and most interesting megatrends in the automotive industry. As we discussed in a previous blog, the reason that this new architectural—and business—model will be successful is the advantages it offers to all stakeholders:
The OEMs (car manufacturers) will gain new revenue streams from aftermarket services and new applications;
The car owners will easily upgrade their vehicle features and functions; and
The mobile operators will profit from increased vehicle data consumption driven by new applications.
What is a software-defined vehicle? While there is no official definition, the term reflects the change in the way software is being used in vehicle design to enable flexibility and extensibility. To better understand the software-defined vehicle, it helps to first examine the current approach.
Today’s embedded control units (ECUs) that manage car functions do include software, however, the software in each ECU is often incompatible with and isolated from other modules. When updates are required, the vehicle owner must visit the dealer service center, which inconveniences the owner and is costly for the manufacturer.
By Kishore Atreya, Director of Product Management, Marvell
Recently the Linux Foundation hosted its annual ONE Summit for open networking, edge projects and solutions. For the first time, this year’s event included a “mini-summit” for SONiC, an open source networking operating system targeted for data center applications that’s been widely adopted by cloud customers. A variety of industry members gave presentations, including Marvell’s very own Vijay Vyas Mohan, who presented on the topic of Extensible Platform Serdes Libraries. In addition, the SONiC mini-summit included a hackathon to motivate users and developers to innovate new ways to solve customer problems.
So, what could we hack?
At Marvell, we believe that SONiC has utility not only for the data center, but to enable solutions that span from edge to cloud. Because it’s a data center NOS, SONiC is not optimized for edge use cases. It requires an expensive bill of materials to run, including a powerful CPU, a minimum of 8 to 16GB DDR, and an SSD. In the data center environment, these HW resources contribute less to the BOM cost than do the optics and switch ASIC. However, for edge use cases with 1G to 10G interfaces, the cost of the processor complex, primarily driven by the NOS, can be a much more significant contributor to overall system cost. For edge disaggregation with SONiC to be viable, the hardware cost needs to be comparable to that of a typical OEM-based solution. Today, that’s not possible.
By Nishant Lodha, Director of Product Marketing – Emerging Technologies, Marvell
While age is just a number and so is new speed for Fibre Channel (FC), the number itself is often irrelevant and it’s the maturity that matters – kind of like a bottle of wine! Today as we make a toast to the data center and pop open (announce) the Marvell® QLogic® 2870 Series 64G Fibre Channel HBAs, take a glass and sip into its maturity to find notes of trust and reliability alongside of operational simplicity, in-depth visibility, and consistent performance.
Big words on the label? I will let you be the sommelier as you work through your glass and my writings.
By Gary Kotzur, CTO, Storage Products Group, Marvell and Jon Haswell, SVP, Firmware, Marvel
The nature of storage is changing much more rapidly than it ever has historically. This evolution is being driven by expanding amounts of enterprise data and the inexorable need for greater flexibility and scale to meet ever-higher performance demands.
If you look back 10 or 20 years, there used to be a one-size-fits-all approach to storage. Today, however, there is the public cloud, the private cloud, and the hybrid cloud, which is a combination of both. All these clouds have different storage and infrastructure requirements. What’s more, the data center infrastructure of every hyperscaler and cloud provider is architecturally different and is moving towards a more composable architecture. All of this is driving the need for highly customized cloud storage solutions as well as demanding the need for a comparable solution in the memory domain.
By Amit Sanyal, Senior Director, Product Marketing, Marvell
SONiC (Software for Open Networking in the Cloud) has steadily gained momentum as a cloud-scale network operating system (NOS) by offering a community-driven approach to NOS innovation. In fact, 650 Group predicts that revenue for SONiC hardware, controllers and OSs will grow from around US$2 billion today to around US$4.5 billion by 2025.
Those using it know that the SONiC open-source framework shortens software development cycles; and SONiC’s Switch Abstraction Interface (SAI) provides ease of porting and a homogeneous edge-to-cloud experience for data center operators. It also speeds time-to-market for OEMs bringing new systems to the market.
The bottom line: more choice is good when it comes to building disaggregated networking hardware optimized for the cloud. Over recent years, SONiC-using cloud customers have benefited from consistent user experience, unified automation, and software portability across switch platforms, at scale.
As the utility of SONiC has become evident, other applications are lining up to benefit from this open-source ecosystem.
A SONiC Buffet: Extending SONiC to Storage
SONiC capabilities in Marvell’s cloud-optimized switch silicon include high availability (HA) features, RDMA over converged ethernet (RoCE), low latency, and advanced telemetry. All these features are required to run robust storage networks.
Here’s one use case: EBOF. The capabilities above form the foundation of Marvell’s Ethernet-Bunch-of-Flash (EBOF) storage architecture. The new EBOF architecture addresses the non-storage bottlenecks that constrain the performance of the traditional Just-a-Bunch-of-Flash (JBOF) architecture it replaces-by disaggregating storage from compute.
EBOF architecture replaces the bottleneck components found in JBOF - CPUs, DRAM and SmartNICs - with an Ethernet switch, and it’s here that SONiC is added to the plate. Marvell has, for the first time, applied SONiC to storage, specifically for services enablement, including the NVMeoFTM (NVM Express over Fabrics) discovery controller, and out-of-band management for EBOF, using Redfish® management. This implementation is in production today on the Ingrasys ES2000 EBOF storage solution. (For more on this topic, check out this, this, and this.)
Marvell has now extended SONiC NOS to enable storage services, thus bringing the benefits of disaggregated open networking to the storage domain.
OK, tasty enough, but what about compute?
How Would You Like Your Arm Prepared?
I prefer Arm for my control plane processing, you say. Why can’t I manage those switch-based processors using SONiC, too, you ask? You’re in luck. For the first time, SONiC is the OS for Arm-based, embedded control plane processors, specifically the control plane processors found on Marvell® Prestera® switches. SONiC-enabled Arm processing allows SONiC to run on lower-cost 1G systems, reducing the bill-of-materials, power, and total cost of ownership for both management and access switches.
In addition to embedded processors, with the OCTEON® family, Marvell offers a smorgasbord of Arm-based processors. These can be paired with Marvell switches to bring the benefits of the Arm ecosystem to networking, including Data Processing Units (DPUs) and SmartNICs.
By combining SONiC with Arm processors, we’re setting the table for the broad Arm software ecosystem - which will develop applications for SONiC that can benefit both cloud and enterprise customers.
The Third Course
So, you’ve made it through the SONiC-enabled switching and on-chip control processing courses but there’s something more you need to round out the meal. Something to reap the full benefit of your SONiC experience. PHY, of course. Whether your taste runs to copper or optical mediums; PAM or coherent modulation, Marvell provides a complete SONiC-enabled portfolio by offering SONiC with our (not baked) Alaska® Ethernet PHYs and optical modules built using Marvell DSPs. Room for Dessert?
Finally, by enabling SONiC across the data center and enterprise switch portfolio we’re able to bring operators the enhanced telemetry and visibility capabilities that are so critical to effective service-level validation and troubleshooting. For more information on Marvell telemetry capabilities, check out this short video:
The Drive Home
Disaggregation has lowered the barrier-to-entry for market participants - unleashing new innovations from myriad hardware and software suppliers. By making use of SONiC, network designers can readily design and build disaggregated data center and enterprise networks.
For its part, Marvell’s goal is simple: help realize the vision of an open-source standardized network operating system and accelerate its adoption.
By Kristin Hehir, Senior Manager, PR and Marketing, Marvell
Flash Memory Summit (FMS), the industry’s largest conference featuring data storage and memory technology solutions, presented its 2022 Best of Show Awards at a ceremony held in conjunction with this week’s event. Marvell was named a winner alongside Exascend for the collaboration of Marvell’s edge and client SSD controller with Exascend’s high-performance memory card.
Honored as the “Most Innovative Flash Memory Consumer Application,” the Exascend Nitro CFexpress card powered by Marvell’s PCIe® Gen 4, 4-NAND channel 88SS1321 SSD controller enables digital storage of ultraHD video and photos in extreme temperature environments where ruggedness, endurance and reliability are critical. The Nitro CFexpress card is unique in controller, hardware and firmware architecture in that it combines Marvell’s 12nm process node, low-power, compact form factor SSD controller with Exascend’s innovative hardware design and Adaptive Thermal Control™ technology.
The Nitro card is the highest capacity VPG 400 CFexpress card on the market, with up to 1 TB of storage, and is certified VPG400 by the CompactFlash® Association using its stringent Video Performance Guarantee Profile 4 (VPG400) qualification. Marvell’s 88SS1321 controller helps drive the Nitro card’s 1,850 MB/s of sustained read and 1,700 MB/sustained write for ultimate performance.
“Consumer applications, such as high-definition photography and video capture using professional photography and cinema cameras, require the highest performance from their storage solution. They also require the reliability to address the dynamics of extreme environmental conditions, both indoors and outdoors,” said Jay Kramer, Chairman of the Awards Program and President of Network Storage Advisors Inc. “We are proud to recognize the collaboration of Marvell’s SSD controllers with Exascend’s memory cards, delivering 1,850 MB/s of sustained read and 1,700 MB/s sustained write for ultimate performance addressing the most extreme consumer workloads. Additionally, Exascend’s Adaptive Thermal Control™ technology provides an IP67 certified environmental hardening that is dustproof, water resistant and tackles the issue of overheating and thermal throttling.”
More information on the 2022 Flash Memory Summit Best of Show Award Winners can be found here.
By Khurram Malik, Senior Manager, Technical Marketing, Marvell
A massive amount of data is being generated at the edge, data center and in the cloud, driving scale out Software-Defined Storage (SDS) which, in turn, is enabling the industry to modernize data centers for large scale deployments. Ceph is an open-source, distributed object storage and massively scalable SDS platform, contributed to by a wide range of major high-performance computing (HPC) and storage vendors. Ceph BlueStore back-end storage removes the Ceph cluster performance bottleneck, allowing users to store objects directly on raw block devices and bypass the file system layer, which is specifically critical in boosting the adoption of NVMe SSDs in the Ceph cluster. Ceph cluster with EBOF provides a scalable, high-performance and cost-optimized solution and is a perfect use case for many HPC applications. Traditional data storage technology leverages special-purpose compute, networking, and storage hardware to optimize performance and requires proprietary software for management and administration. As a result, IT organizations neither scale-out nor make it feasible to deploy petabyte or exabyte data storage from a CAPEX and OPEX perspective. Ingrasys (subsidiary of Foxconn) is collaborating with Marvell to introduce an Ethernet Bunch of Flash (EBOF) storage solution which truly enables scale-out architecture for data center deployments. EBOF architecture disaggregates storage from compute and provides limitless scalability, better utilization of NVMe SSDs, and deploys single-ported NVMe SSDs in a high-availability configuration on an enclosure level with no single point of failure.
Ceph is deployed on commodity hardware and built on multi-petabyte storage clusters. It is highly flexible due to its distributed nature. EBOF use in a Ceph cluster enables added storage capacity to scale up and scale out at an optimized cost and facilitates high-bandwidth utilization of SSDs. A typical rack-level Ceph solution includes a networking switch for client, and cluster connectivity; a minimum of 3 monitor nodes per cluster for high availability and resiliency; and Object Storage Daemon (OSD) host for data storage, replication, and data recovery operations. Traditionally, Ceph recommends 3 replicas at a minimum to distribute copies of the data and assure that the copies are stored on different storage nodes for replication, but this results in lower usable capacity and consumes higher bandwidth. Another challenge is that data redundancy and replication are compute-intensive and add significant latency. To overcome all these challenges, Ingrasys has introduced a more efficient Ceph cluster rack developed with management software – Ingrasys Composable Disaggregate Infrastructure (CDI) Director.
While there are
32GB micro-SD or USB boot device options available today —
with VMware requiring as much as 128GB of storage for the OS and Microsoft
Storage Spaces Direct needing 200GB — these solutions simply don’t have the
storage capacity needed. Using hardware RAID controllers and disk drives in the
server bays is another option. However, this adds significant cost and
complexity to a server configuration just to meet the OS requirement. The
proper solution to address separating the OS from user data is the HPE NS204i-p NVME OS Boot Device.
Flash Memory Summit, the industry's largest trade show dedicated to flash memory and solid-state storage technology, presented its 2020 Best of Show Awards yesterday in a virtual ceremony. Marvell, alongside Hewlett Packard Enterprise (HPE), was named a winner for "Most Innovative Flash Memory Technology" in the controller/system category for the Marvell NVMe RAID accelerator in the HPE OS Boot Device.
Last month, Marvell introduced the industry’s first native NVMe RAID 1 accelerator, a state-of-the-art technology for virtualized, multi-tenant cloud and enterprise data center environments which demand optimized reliability, efficiency, and performance. HPE is the first of Marvell's partners to support the new accelerator in the HPE NS204i-p NVMe OS Boot Device offered on select HPE ProLiant servers and HPE Apollo systems. The solution lowers data center total cost of ownership (TCO) by offloading RAID 1 processing from costly and precious server CPU resources, maximizing application processing performance.
You are an avid gamer. You spend countless hours in forums to decide between the ASUS TUF components and researching Radeon RX 500 or GeForce RTX 20, to ensure games would show at their best on your hard-earned PC gaming rig. You made your selection and can’t stop bragging about your system’s ray tracing capabilities and how realistic is the “Forza Motorsport 7” view from your McLaren F1 GT cockpit when you drive through the legendary Le Mans circuit at dusk. You are very proud of your machine and the year 2020 is turning out to be good: Microsoft finally launched the gorgeous looking “Flight Simulator 2020,” and CD Projekt just announced that the beloved and award-winning “The Witcher 3” is about to get an upgrade to take advantage of the myriad of hardware updates available to serious gamers like you. You have your dream system in hand and life can’t be better.
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.
In this VMware BLOG, the author outlines that server elements in the bom can change including:
CPU
存储器
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 supportUniversal RDMA – the ability to support both RoCE and IWARP RDMA protocols with the same adapter.
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.
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.
One of the questions we get quite often is which protocol is best for connecting servers to shared storage?
What customers need to do is look at the reality of what they need from a shared storage environment and make a decision based on cost, performance and manageability.
Notes: 1. Optical transceiver needed at both adapter and switch ports for 10GbE networks. Thus cost/port is two times the transceiver cost 2. FC switch pricing includes full featured management software and licenses 3. FC Host Bus Adapters (HBAs) ship with transceivers, thus only one additional transceiver is needed for the switch port
So if we do the math, the cost per port looks like this:
10GbE iSCSI with SFP+ Optics = $437+$2,734+$300 = $3,471
10GbE iSCSI with 3 meter Direct Attach Cable (DAC) =$437+$269+300 = $1,006
Note, in my example, I chose 3 meter cable length, but even if you choose shorter or longer cables (HPE supports from 0.65 to 7 meter cable lengths), this is still the lowest cost connection option. Surprisingly, the cost of the 10GbE optics makes the iSCSI solution with optical connections the most expensive configuration.
It really comes down to distance and the number of connections required. The DAC cables can only span up to 7 meters or so. That means customers have only limited reach within or across racks. If customers have multiple racks or distance requirements of more than 7 meters, FC becomes the more attractive option, from a cost perspective.
Latency is an order of magnitude lower for FC compared to iSCSI. Latency of Brocade Gen 5 (16Gb) FC switching (using cut-through switch architecture) is in the 700 nanoseconds range and for 10GbE it is in the range of 5 to 50 microseconds. The impact of latency gets compounded with iSCSI should the user implement 10GBASE-T connections in the iSCSI adapters. This adds another significant hit to the latency equation for iSCSI.
Keep in mind, Ethernet network management hasn’t really changed much. Network administrators create virtual LANs (vLANs) to separate network traffic and reduce congestion. These network administrators have a variety of tools and processes that allow them to monitor network traffic, run diagnostics and make changes on the fly when congestion starts to impact application performance.
On the FC side, companies like Cavium and HPE have made significant improvements on the software side of things to simplify SAN deployment, orchestration and management. Technologies like fabric-assigned port worldwide name (FA-WWN) from Cavium and Brocade enable the SAN administrator to configure the SAN without having HBAs available and allow a failed server to be replaced without having to reconfigure the SAN fabric. Cavium and Brocade have also teamed up to improve the FC SAN diagnostics capability with Gen 5 (16Gb) Fibre Channel fabrics by implementing features such as Brocade ClearLink™ diagnostics, Fibre Chanel Ping (FC ping) and Fibre Channel traceroute (FC traceroute), link cable beacon (LCB) technology and more. HPE’s Smart SAN for HPE 3PAR provides the storage administrator the ability to zone the fabric and map the servers and LUNs to an HPE 3PAR StoreServ array from the HPE 3PAR StoreServ management console.
In many enterprise environments, there are typically dozens of network administrators. In those same environments, there may be less than a handful of “SAN” administrators. Yes, there are lots of LAN connected devices that need to be managed and monitored, but so few for SAN connected devices.
Well, it depends. If application performance is the biggest criteria, it’s hard to beat the combination of bandwidth, IOPS and latency of the 16GFC SAN. If compatibility and commonality with existing infrastructures is a critical requirement, 10GbE iSCSI is a good option (assuming the 10GbE infrastructure exists in the first place). If security is a key concern, FC is the best choice. When is the last time you heard of a FC network being hacked into?
因此，为了能够控制开销，尽可能提高运营效率，数据中心和商业企业将尝试采用分层存储方法，使用非常合适的存储介质，以降低相关成本。 存储介质的选择将依据数据访问的频率及可接受的延迟程度而定。 This will require the use of numerous different technologies to make it fully economically viable - with cost and performance being important factors.
全球科技领域的主要代表上周齐聚圣何塞会展中心，共同谛听今年年度电子创新 (ACE) 大奖花落谁家。 此项久负盛名的赛事由杰出的电子工程杂志 EDN 和 EE Times 共同组织，重点奖励过往 12 个月推向市场的非常具有创新品质的产品，以及非常具有前瞻意识的高层领导和非常有潜力的创业公司。 由杂志的编辑团队和多位备受尊敬的独立评委组成的评审团，将参与选择每个分类的大奖得主。Marvell 于今年早些时候推出的 88NV1160 高性能控制器用于非易失性存储器 (NVMe)，该产品击败 Diodes Inc. 和 Cypress 半导体公司等生产的强劲竞争产品，赢得梦寐以求的逻辑/接口/内存类大奖。 Marvell gained two further nominations at the awards - with 98PX1012 Prestera PX Passive Intelligent Port Extender (PIPE) also being featured in the Logic/Interface/Memory category, while the 88W8987xA automotive wireless combo SoC was among those cited in the Automotive category.
通过与主机存储缓冲技术的配合，88NV1160 表现出大大低于同类竞争产品的延迟率。 正是这一点支持该设备极大提高了读取速度。 By utilizing its embedded SRAM, the controller does not need to rely on an external DRAM memory - thereby simplifying the memory controller implementation. 并最终大幅减少所需的主板空间，同时降低了产品的总体物料清单成本。
88NV1160 特有的 NANDEdge™ 低密度奇偶校验纠错功能提高了 SSD 的耐久度，确保最终产品在整个使用寿命期间，都能维持长期的系统可靠性。 该控制器内置 256 位 AES 加密引擎，能够保证所存储元数据的安全性，使其不因潜在的安全漏洞而遭到攻击。 除此之外，这些无 DRAM 的 IC 外形非常紧凑，因而非常有利于实现多芯片封装的集成。
现在客户都希望他们的便携式电子设备能够具备更多的计算资源，这样才可以使用大量令人垂涎的新上市软件应用；充分利用基于云的服务，享受增强现实和游戏的乐趣。 在提供此类功能的同时，此类设备产品需要能够支持更长的电池使用时间，减少充电次数，如此方能进一步提高用户所得到的使用体验。 客户需要先进的 IC 在具备强大处理能力的同时改善能耗水平，而 88NV1160 正好满足了这种需求。 "We're excited to honor this robust group for their dedication to their craft and efforts in bettering the industry for years to come," said Nina Brown, Vice President of Events at UBM Americas. "The judging panel was given the difficult task of selecting winners from an incredibly talented group of finalists and we'd like to thank all of those participants for their amazing work and also honor their achievements. These awards aim to shine a light on the best in today's electronics realm and this group is the perfect example of excellence within both an important and complex industry."
SATA 仍然是当前最常见的存储接口。 无论是硬盘驱动器，还是日益盛行的闪存存储设备，大部分都是采用 SATA 接口来工作。 The latest generation of SATA - SATA III – has a 600 MB/s bandwidth limit. 虽然这个带宽足已满足日常消费类应用的需求，但对于企业服务器而言，还是远远不够。 即便是在 I/O 密集型的消费类应用场景下（例如视频编辑）也可能受限于这个带宽。
SATA 标准最初在 2000 年作为旧式 PATA 并行接口标准的串行替代标准而发布。 SATA 采用高级主机控制接口 (AHCI) 技术，单个命令队列的深度为 32 个命令。 此命令队列架构非常适合传统的磁碟式硬盘存储，但用于闪存时，却存在较多限制。
如果说 SATA 是消费类驱动器的标准存储接口，那么 SAS 则在企业应用中更为常见。 Released originally in 2004, SAS is also a serial replacement to an older parallel standard SCSI. Designed for enterprise applications, SAS storage is usually more expensive to implement than SATA, but it has significant advantages over SATA for data center use - such as longer cable lengths, multipath IO, and better error reporting. SAS 的带宽限制也更高，为 1200MB/s。
尽管 SAS 支持的队列深度为 254 个命令，大大多于 SATA 的 32 个，但和 SATA 相同，SAS 依然只支持单个命令队列。 虽然 SAS 具有更深的命令队列和更高的带宽限制，总体性能优于 SATA，但也远非理想的闪存接口。
NVMe - Massive Parallelism
NVMe 于 2011 年推出，从设计之初就充分考虑到闪存存储的需求。 NVMe 由多家存储公司联合开发，其主要目标是克服 SATA 和 SAS 对闪存性能造成的瓶颈。
企业存储通常由数据中心多层机架内互相连通的磁盘阵列组成，而嵌入式存储一般只是简单的将一块固态硬盘 (SSD) 安装到嵌入式电脑或设备之中。 企业经常会控制数据中心的物理安全，也会执行企业网络（或应用程序）的软件访问控制。 Embedded devices, on the other hand - such as tablets, industrial computers, smartphones, or medical devices - are often used in the field, in what are comparatively unsecure environments. 这种背景下的数据安全没有其他选择，只能在设备层面进行。
Because software based FDE relies on the host processor to perform encryption, it is usually slower - whereas hardware based FDE has much lower overhead as it can take advantage of the drive’s integrated crypto-processor. 基于硬盘的 FDE 还可以加密硬盘的主引导记录，与之相反，基于软件的加密却不具备这种功能。
When solid-state drives (SSDs) were first introduced into the data center, the infrastructure mandated they work within the confines of the then current bus technology, such as Serial ATA (SATA) and Serial Attached SCSI (SAS), developed for rotational media. 即使非常快的 HDD 也比不上 SSD 的速度，其对应总线的吞吐能力也成为阻碍充分发挥 SSD 技术优势的一大瓶颈。 作为一种在网络，图形以及其他插入式设备上广泛应用的高带宽总线， PCIe 成为了可行的选择，但 PCIe 总线配合原本为 HDD 开发的存储协议（例如 AHCI）仍然无法有效发挥易失性存储介质的性能优势。 因此，NVM Express (NVMe) 行业工作组成立，创建一组为 PCIe 总线开发的标准化协议和命令，以便允许多条路径可以充分利用数据中心中 SSD 的优势。 The NVMe specification was designed from the ground up to deliver high-bandwidth and low-latency storage access for current and future NVM technologies.
New NVMe controllers from companies like Marvell allowed the data center to share storage data to further maximize cost and performance efficiencies. 使用 SSD 建立存储集群，取代以往为每台服务器单独配置存储的方式，提高数据中心存储总容量。 此外，通过为其他服务器建立公共区传送数据，可以方便地访问共享数据。 因此，这些新计算模型使数据中心不仅能够充分优化 SSD 的高速性能，而且能够更经济地在整个数据中心部署 SSD，从而降低总体成本并简化维护。 针对负载较高的服务器，不用增加额外的 SSD，而是从存储池动态分配满足其需求。
The Marvell 88SS1092 is Marvell's second-generation NVMe SSD controller capable of PCIe 3.0 X 4 end points to provide full 4GB/s interface to the host and help remove performance bottlenecks. While the new controller advances a solid-state storage system to a more fully flash-optimized architecture for greater performance, it also includes Marvell's third-generation error-correcting, low-density parity check (LDPC) technology for the additional reliability enhancement, endurance boost and TLC NAND device support on top of MLC NAND.
We pride ourselves on delivering innovative solutions to help our global customers store, move and access data—fast, securely, reliably and efficiently. In further recognition of our world-class technology, we are excited to share that The Linley Group, one of the most prominent semiconductor analyst firms, has selected Marvell’s ARMADA® SP (storage processor) as the Best Embedded Processor in its 2016 Analysts' Choice Awards.
Honoring the best and the brightest in semiconductor technology, the Analysts' Choice Awards recognize the solutions that deliver superior power, performance, features and pricing for their respective end applications and markets.
To learn more about Marvell’s SP solution, visit: http://www.marvell.com/storage/armada-sp/.
Powered by the Marvell® Bravera™ SC5 controller, Memblaze developed the PBlaze 7 7940 GEN5 SSD family, delivering an impressive 2.5 times the performance and 1.5 times the power efficiency compared to conventional PCIe 4.0 SSDs and ~55/9us read/write latency1. This makes the SSD ideal for business-critical applications and high-performance workloads like machine learning and cloud computing. In addition, Memblaze utilized the innovative sustainability features of Marvell’s Bravera SC5 controllers for greater resource efficiency, reduced environmental impact and streamlined development efforts and inventory management.
HPE GreenLake for Block Storage powered by Alletra Storage MP
Room for Dessert?
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.
Notes: 1. Optical transceiver needed at both adapter and switch ports for 10GbE networks. Thus cost/port is two times the transceiver cost 2. FC switch pricing includes full featured management software and licenses 3. FC Host Bus Adapters (HBAs) ship with transceivers, thus only one additional transceiver is needed for the switch port 
欲了解有关 Marvell 产品如何存储全球数据的更多信息，请访问 www.marvell.com。
全球科技领域的主要代表上周齐聚圣何塞会展中心，共同谛听今年年度电子创新 (ACE) 大奖花落谁家。 此项久负盛名的赛事由杰出的电子工程杂志 EDN 和 EE Times 共同组织，重点奖励过往 12 个月推向市场的非常具有创新品质的产品，以及非常具有前瞻意识的高层领导和非常有潜力的创业公司。 由杂志的编辑团队和多位备受尊敬的独立评委组成的评审团，将参与选择每个分类的大奖得主。
Marvell 于今年早些时候推出的 88NV1160 高性能控制器用于非易失性存储器 (NVMe)，该产品击败 Diodes Inc. 和 Cypress 半导体公司等生产的强劲竞争产品，赢得梦寐以求的逻辑/接口/内存类大奖。 Marvell gained two further nominations at the awards - with 98PX1012 Prestera PX Passive Intelligent Port Extender (PIPE) also being featured in the Logic/Interface/Memory category, while the 88W8987xA automotive wireless combo SoC was among those cited in the Automotive category. 
"We're excited to honor this robust group for their dedication to their craft and efforts in bettering the industry for years to come," said Nina Brown, Vice President of Events at UBM Americas. "The judging panel was given the difficult task of selecting winners from an incredibly talented group of finalists and we'd like to thank all of those participants for their amazing work and also honor their achievements. These awards aim to shine a light on the best in today's electronics realm and this group is the perfect example of excellence within both an important and complex industry." 
数据中心 SSD： NVMe and Where We’ve Been 
We pride ourselves on delivering innovative solutions to help our global customers store, move and access data—fast, securely, reliably and efficiently. In further recognition of our world-class technology, we are excited to share that The Linley Group, one of the most prominent semiconductor analyst firms, has selected Marvell’s ARMADA® SP (storage processor) as the Best Embedded Processor in its 2016 Analysts' Choice Awards. 
