Today’s storage area networks (SANs) face tremendous pressure from the phenomenal growth of digital information and the need to access it quickly and efficiently. Worldwide data is projected to multiply by an astonishing 1000 percent by 2020. It’s little wonder, then, that storage administrators rank slow drain and related SAN congestion issues as their number-one concern. If not addressed in a timely fashion, these can have a domino effect, even degrading the performance of totally unrelated applications.
A Slow Drain Device is a device that does not accept frames at the rate generated by the source. In the presence of slow devices, Fibre Channel networks are likely to lack frame buffers, resulting in switch port credit starvation and potentially choking Inter-Switch Links (ISLs). Frames destined for slow devices need to be carefully isolated in separate queues and switched to egress ports without congesting the backplane. A decision then needs to be made about whether the frames are considered stuck and when to drop them.
Cisco provides a Slow Drain Device Detection and Congestion Avoidance (referred to Slow Drain) feature that helps detect, identify, and resolve the condition exhibited by slow devices.
Join us for this live 60-minute webcast and learn the common causes of slow drain and other typical SAN congestion issues. See how Cisco Nexus and MDS switches now include hardware-based congestion detection and recovery logic for precise, fast detection and automatic, real-time resolution.
Watch this Video to learn more: This video demonstrates slow drain diagnostics using Cisco Prime DCNM . See how DCNM can be used to detect Slow Drain devices and troubleshoot the situation within minutes in a large fabric with thousands of ports. Deploy Cisco Prime DCNM today to bring down Slow Drain troubleshooting time from day or weeks to minutes.
Reasons for Slow Drain include:
1) An edge device can be slow to respond for a variety of reasons: Server performance problems: application or OS, Host bus adapter (HBA) problems: driver or physical failure , Speed mismatches: one fast device and one slow device , Nongraceful virtual machine exit on a virtualized server, resulting in packets held in HBA buffers , Storage subsystem performance problems, including overload, Poorly performing tape drives
2) Inter Switch Links (ISL) can be slow due to: Lack of B2B credits for the distance the ISL is traversing and the existence of slow drain edge devices ( in the fabric)
Any device exhibiting such behavior is called a Slow Drain Device. Cisco MDS 9000 Family switches constantly monitor the network for symptoms of slow drain and can send alerts and takes automatic actions to mitigate the situation. Read this whitepaper to learn more.
Dell’Oro just released the Q3 CY2014 SAN Switching Market Share results, and for the second quarter in a row, Cisco came out at #1 in combined FC + FCoE revenue. Only this time, we widened the gap by 9 points! We also took #1 in FC Modular share, with an impressive 11.6% Q/Q share gain. Our FC Fixed share has been growing too, up 8% from a year ago. In all, Cisco has really stepped up its game in SAN Switching.
Cisco market share gains are primarily due to customer acceptance of the MDS product portfolio, ramp up of FCoE-capable products, along with substantial 16G FC demand. FCoE is supported on MDS, Nexus and UCS providing an end to end architecture enabling customers to reduce cost and simplify operations. In the recent quarters, customer buying trends also shifted from buying point products to acquiring pre-packaged solutions. NX-OS a single operating system, and DCNM as single management platform to manage across LAN and SAN, resonated well with customers. Cisco‘s architectural approach of combining storage, network, compute and management helps to reduce cost and increase efficiency.
Cisco Storage Networking – Performance, Flexibility and Availability
Cisco Storage Innovation: Over the last year, Cisco has made significant investments in both FC and FCoE on the MDS and Nexus product lines, and we are starting to see those investments pay off. Last year Cisco introduced a 16G Director (MDS 9710), offering 3x the performance of our competitors’ and also a new Multi Service Platform (MDS 9250i), which we like to refer to as “the Swiss Army Knife” of SAN switching, due to its multi-protocol (16G FC, 10G FCoE, and 10G FCIP ports) and multi-service (SAN Extension, IO Acceleration, and Data Mobility Migration) capabilities. Three months ago we introduced a new compact director (MDS 9706) and a 16G Fabric Switch (MDS 9148S), along with new FCoE enhancements providing customers with deployment choices.
Data centers are undergoing a major transition to meet higher performance, scalability, and resiliency requirements with fewer resources, smaller footprint, and simplified designs. These rigorous requirements coupled with major data center trends, such as virtualization, data center consolidation and data growth, are putting a tremendous amount of strain on the existing infrastructure and adding complexity. Designing a reliable, scalable, and flexible data center requires careful planning.
Register for this webcast to learn about the Cisco MDS 9710 Multilayer Director which power high performance, flexible and scalable data center design. We will look at some of the data center architecture options that come with the Cisco MDS 9710 Multilayer Director as well as review the best practices to improve resiliency and protect against silent failures. We will also discuss how we can design a data center that provides investment protection and agility without making tradeoffs.
A couple of months ago, we introduced the new Cisco MDS 9700 Series Multilayer Directors. The new Cisco MDS 9710 Director establishes a new benchmark for performance – delivering more than three times the bandwidth of any Director in the industry. Since the launch, several of our customers deployed MDS 9710 successfully and reaped the benefit from it.
In this blog, let me share with you on how LinkedIn used Cisco MDS to solve their storage growth needs. Since the company was founded, it has enjoyed steady growth in members and revenues. In fact, based on LinkedIn’s calculations, LinkedIn adds more than two new members per second. This means that the company’s storage needs and storage area network (SAN) are growing fast, and its SAN switches need to keep pace. With more than 259 million registered members, LinkedIn is connecting the world’s professionals to make them more productive and successful.
For about 2 decades, Fibre Channel has been deployed to support the most rigorous and demanding applications, and is mainly because FC possesses certain characteristics that enterprise customers embrace, i.e – reliability, performance and scalability. 16Gb Fibre Channel is one of the preferred choice for mission-critical applications that required deterministic low latency networks to connect to storage. In addition to emerging technologies like FCoE and 40 GE, enterprise customers still continue to deploy Fibre channel for mission critical applications and this trend will continue for a while, due to huge customer investment in FC infrastructure and how their IT departments are organized. According to Dell’Oro report (June, 2013), 38% of data center switch/director ports are projected to be 16 Gb FC during 2014.
In order to support this demand for 16G, you may expect more product updates and new offerings from Fibre Channel vendors. For example, EMC announced their new 16G FC storage networking connectivity that further extends EMC’s lead in the market. Customers can now take advantage and future-proof their virtual data centers by deploying complete, end-to-end 16 Gb Fibre Channel connectivity throughout the cloud. A couple of months ago Cisco announced high density 16G and FCoE platform –Cisco MDS 9700 director , which delivers superior performance—more than three times the bandwidth than any currently available SAN Director in the market.
Well, I want to be very clear, 16G is one of several choices, when you are ready for convergence then you will have a choice of platforms that will support FC, FCoE and/or iSCSI, or all three! You can future proof your storage infrastructure!!! (see figure below). Cisco MDS 9700 (only director that supports both FC and FCOE in the same platform) + EMC solution enable customers to deploy Teir 1, 16Gb storage infrastructure and scale efficiently.