Cisco announces new Service Oriented SAN capabilities

This Silverton Consulting (SCI) Storage Intelligence (StorInt™) Dispatch provides a summary of Cisco’s recent announcement expanding their service oriented SAN technology and offerings.

Service oriented SAN

Sometime back (1.5 years ago) Cisco announced some new line cards for their MDS 9222i SAN directors which provided storage media encryption (SME) and data mobility migration (DMM) services for their SANs.  In conjunction with Cisco SAN packet forwarding capabilities, these boards could be attached anywhere in your Cisco SAN network and support these services for any defined SAN link.  With this announcement, Cisco’s Service Oriented SANs have now been expanded and are now also available in a standalone hardware appliance.

SSN-16

Before the Storage Services Node (SSN-16), the only way to provide Cisco SAN services such as these was to already have a MDS 9500 Director and purchase the appropriate 9222i line card.  With the SSN-16, Cisco now offers a multi-service module that can be software configured to support any 4 of these services (see below). The SNS-16 supports 16-FC ports and can support up to 4 software configured, service engines.  The SNS-16 also comes with 2 GigE ports for management access.  Additionally, the data migration service is available in a separate standalone appliance, to provide greater flexibility for enabling data migration/upgrade projects.

Just like the MDS-9222i service modules, the SSN-16 can be configured to support services for any defined SAN link.  Once installed in your SAN, configuration and enabling/disabling of SAN service modules should be non-disruptive.

I/O Accelerator

The new I/O Accelerator services optimize long distance data transfers, such as backups or array-based replication.  Such capabilities were previously only delivered as isolated , dedicated devices at both ends of a MAN / WAN pipe.  All of these capabilities and more can now be had when deployed as a pair of Cisco I/O Accelerator services on Cisco SSN-16 or other modules within the MDS platform.

The I/O accelerator service accelerates write activity by buffering write data locally, acknowledging write completion early and transferring the data later.  In addition, the service module can provide read-ahead buffering of data.  So for sequential reading and writing to remote locations over the SAN, an I/O Accelerator can speed up I/O activity considerably.

Secure Erase

As storage undergoes technology refresh, it often is retired and sold off or returned as it comes off lease.  In the old days, unless one crushed the embedded disk devices (reducing retention value to almost $0) or performed a multi-day secure erase procedure, your data could be readable by the next company to purchase your storage hardware.

With Cisco’s new Secure Erase service, all one needs to do is to activate the service to securely erase a bunch of LUNs and stand back.  Cisco will take the erase workload off your hosts and place it in the SAN.  Cisco states that they support MilSpec Data Security Erase standards which include multi-pass overwrites with periodic data patterns to verifiably obscure and erase any data previously present.

Announcement significance

Cisco Service-Oriented SAN offerings are starting to be made available to service providers and other channels to be used in tech refresh operations.  As such, a secure erase service and a standalone appliance make a lot of sense.  Overall, Cisco seems to be using service modules as a way to differentiate their SAN offerings.

As for performance, SME induces some small latency to every tape write or reads going through it.   The I/O Accelerator looks to be nothing more than an I/O cache in the network between remote links and as such, can only improve remote access response times.  For DMM, the fact that migrations can be done online while continuing to access the old data means that they are no longer confined to night and weekend outage windows, so  the migration can be accomplished over a much shorter duration.  Secure erase takes days any way you cut it, but taking the burden off the host can only help.

A PDF version of this can be found at

Cisco 2010 January 26 Announcement of New Service Oriented SAN capabilities (PDF 255.2 KiB)

Silverton Consulting, Inc. is a Storage, Strategy & Systems consulting services company, based in the USA offering products and services to the data storage community.

 

Announcing Brocade DCX-4S

This Silverton Consulting (SCI) Storage Intelligence (StorInt™) Dispatch provides a summary of Brocade’s recent announcement of their DCX-4S and other items.

DCX-4S

Brocade is enjoying significant success with their 8Gb fibre channel DCX Backbone, first released in January 2008 and proving to be its fastest ramping modular switch in Brocade’s history. Their newest addition to this product family is the DCX-4S, which is a four horizontal slot version of the bigger DCX model, but with full backbone-class performance, energy efficiency, and advanced functionality. The DCX-4S also offers high-speed Inter Chassis Links (ICLs) which can cross link a DCX-4S to its bigger brother, the DCX or to another DCX-4S.  This can be used to provide more scalable and flexible backbone configurations at the network core and edge.

The new DCX-4S can provide half the number of ports (192 FC/FICON ports) and switch throughput (1.536Tb/s) of the larger 384-port DCX model.  Each DCX-4S slot can support a Fibre Channel blade with from 16 to 48 8GFC ports, or an application blade such as the Fabric Encryption blade or the Brocade Application blade for EMC RecoverPoint, which are also supported by the DCX model.  The DCX-4S can connect to Brocade B- and M-Series SAN fabrics without disruption and with common management. And its multiprotocol architecture is designed to support emerging Converged Enhanced Ethernet (CEE) and Fibre Channel over Ethernet (FCoE) protocols through the addition of a future blade.

Fabric Operating System Virtual Fabrics

Brocade is also introducing their Virtual Fabric support in its Fabric Operating System (FOS) that provides for partitioning a physical switch into multiple logical switches.  Virtual Fabrics are very useful for isolating SAN traffic via independent logical switches managed as separate entities.  Some use cases for Virtual Fabrics include for mainframes, isolating FICON from FC traffic, for SAN consolidation, retaining segregated management, and for multi-fabric environments, growth can be managed from a pool of physical ports.

Physical ports can be dynamically allocated to a logical switch offering flexible scalability.  Brocade also mentioned that support for Virtual Fabrics was provided in their 8G FC ASIC and as such, any switch with this ASIC could support Virtual Fabrics. Today, the DCX, DCX-4S, 5300 and 5100 support the feature.  Older products and/or other switches not using the feature can interoperate with Virtual Fabrics by connecting to a single Virtual Fabric.

Fabric level encryption

As for encryption, Brocade also announced support for the HP Secure Key Management appliance and support for IPv6.  Also announced was support for NetVault/BakBone 8.1 and HP Data Protector 6.0 backup applications.  More interesting perhaps, Brocade also announced support for encrypting tape with compression.

HBA announcements

Brocade is announcing support for quality of service (QoS) configurations and SAN boot auto-configuration capabilities for their HBA product line.  Brocade’s HBA is now qualified by EMC, HDS, LSI Corp., and Xiotech and can now be sold by most of them.

HBA port/NPIV-id QoS can be specified to be High, Medium, or Low and be maintained throughout a QoS enabled Brocade switching fabric.  As such, VMs using NPIV-ids can have their port QoS move with their VM as it is Vmotioned throughout a VMware data center.   Such QoS support can limit the cross VM/system performance impact of logical or physical link problems.

SAN boot auto-configuration supports a switch defined automatic configuration of an HBA port at boot time.  Historically, this was maintained in HBA non-volatile memory and as such was somewhat hard to change and error prone. With this new capability, the HBA port boot characteristics are defined at the fabric level and are downloaded whenever a boot request is issued to the HBA.  Current OSs supported by SAN boot auto-configuration includes Windows, Linux, and VMware servers.

Announcement significance

The DCX-4S takes Brocade’s latest switch technology to the mid-market and in the process makes for a much more flexible fabric configuration.  More interesting is their relative success and high adoption rate of the DCX series of products.  Unclear if this reflects the adoption of 8GFC in the enterprise, the relative need for more switch ports/bandwidth, or some need for DCX advanced features.  Most likely a combination of all the above are driving adoption.  How this will play out in the mid market is TBD.

As for the HBA business, one reason Brocade cited for getting into the HBA business was advancing HBA features would require tighter integration with the switching fabric.  However, SCI feels this is more a statement of a lack of standards than a real constraint.

Finally data at rest encryption for tape and disk is now available everywhere, i.e. from the host/server, standalone appliance, fabric switch, storage subsystem or disk/tape device.

A PDF version of this can be found at

Brocade 2009 January 27 Announcement on new DCX-4S switch (PDF 222.0 KiB)

Silverton Consulting, Inc. is a Storage, Strategy & Systems consulting services company, based in the USA offering products and services to the data storage community.

 

Announcing VMware Aware Networking

This Silverton Consulting (SCI) Storage Intelligence (StorInt™) Dispatch provides a summary of Cisco’s recent announcements introducing Virtual Machine aware networking including a new VMware software switch (1000V).

VM networking redefined using VN-Link

VN-Link provides Ethernet, SAN level QOS and other attributes typically defined at the physical server or port level to now be defined down to the VM level. In the past VLANs were defined at the MAC layer and VSANs were defined at the port layer and provided only server level granularity and functionality.  These new VN-Link offerings provide for a much more individualized networking environment that can both be tailored to the VM level and once defined sticks with that VM as its moved throughout the physical servers under VM/HA and VM/DRS control.

Software VN-Link Ethernet networking services – Nexus 1000V

Cisco has introduced the Nexus 1000V as a replacement for VMware’s NIC virtualization (vSwitch) that implements a new capability called a Distributed Virtual Switch.  This is essentially a software Nexus switch residing inside the hypervisor using VMware defined APIs implementing a layer 2 switch.  Nexus 1000V maintains state information within an ESX server and across ESX servers to other 1000Vs without modifying Ethernet packets. As such, the Nexus 1000V can attach to other non-Cisco hardware switches upstream and continue to provide enhanced VM aware service to other Nexus 1000V switches.

Each 1000V software service running in an ESX server is considered as a software line card with an infinite number of ports called a Virtual Ethernet Module (VEM).  A Virtual Supervisor Module (VSM), possibly running as a virtual machine, supplies management, monitoring and configuration of VEMs and is compatible to the hardware version of Nexus Data Center manager. There can be up to 64 VEMs in one distributed virtual switch under VSM control.

The Nexus 1000V runs on the next version of VMware’s ESX server software and will likely be available first half of 2009.  Pricing and licensing have yet to be defined, the Nexus 1000V will most certainly become the value added version of the free vSwitch services available with any ESX server.

Hardware VN-Link Ethernet networking services – Nexus 5000

Cisco is also introducing the hardware Nexus 5000 switch that provides 1000V like services but uses a new Ethernet protocol standard.  The Nexus 5000 uses VM level tagging on network packet headers as defined by the new 802.1Q protocol.  Once packets are tagged with their VM-VLAN associations switches can provide the ability to manage network services at the VM-VLAN level which means that QOS and other networking attributes can be applied to the VM rather than at just the ESX server level.  Now, hardware defined network policies can migrate with VMs as they are moved to other ESX servers.

The new Ethernet 802.1Q standard modifies the standard packet header to add VLAN tagging and was just adopted this past September.  As it’s a protocol change, compatible hardware is required for NICs and switches to support this functionality.  The Nexus 5000 is the first switch to support for this new packet tagging protocol.

Hardware VN-Link storage networking services

In conjunction with the above, Cisco has applied VN-Link to storage networking.   With F-Port trunking from the HBA to the switch port, multiple VMs sharing the same physical HBA can now belong to different VSANs.  This is based on the current ISL level F-port trunking technology.  This new capability is ANSI T11 compliant and requires support at the HBA and the hypervisor level.  Similar to the Ethernet capabilities above, VSAN level QOS and other services will now move around with the VMs as they migrate to other ESX servers.

Other Cisco announcements

New 8GFC cards were announced supporting 24-8GFC, 48-8GFC or a host selectable 4-8GFC/44-4GFC port link cards and are compatible with any MDS 9500 series switch.  Also announced was support for Cisco TrustSec-FC link encryption that provides data-in-flight encryption services for ISL links.  Cisco has rebranded SAN-OS as NX-OS and now both the latest Cisco Ethernet and storage switches run the same O/S.

In addition, Cisco announced their new WAVE product line, providing enhanced VDI acceleration using WAN optimization over new hardware platforms for remote office environments.   With the new WAVE products user login, email, MS office and Web services can approach onsite LAN response times at remote offices.

Cisco also announced additional VMware training and certification to insure partner quality and ability to service these new VM capabilities.  Finally, Cisco announced new professional service offerings, which provide Cisco and VMware consulting in combination to address end-user needs.

These new offerings provide seamless server virtualization assessment, configuration, and ongoing operations support for VMware environments.  Such services can involve Cisco and VMware supplied personnel in any combination, under one contracted service engagement to address customer requirements.

Announcement significance

As server virtualization was being rapidly adopted throughout an enterprise, some networking functionality has been left behind.  Ethernet and storage networking characteristics had always been defined at the server level but this granularity was lost when these same servers were converted to VMs.  VN-Link takes the first steps to move these capabilities back to the VM level allowing QOS, control and management to be returned to where they were prior to virtualization, that is, at the server (or in this case VM) level.  More needs to come in this arena and broader support for other virtualization engines needs to be available but Cisco has taken the first big step in the right direction.

A PDF version of this can be found at

Cisco 2008 September 30 Announcement on VVMware Aware Networking (PDF 319.2 KiB)

Silverton Consulting, Inc. is a Storage, Strategy & Systems consulting services company, based in the USA offering products and services to the data storage community.

 

Announcing:  New Nexus 7000 Series data center class switch

Cisco announced the Nexus 7000 Series switch in January and did an analyst briefing in late February.

Nexus 7000 next generation switch

This is Cisco’s 10Gbe data center core switching architecture supporting unified fabric services for FCoE, IP, HPC and Ethernet. Also, the Nexus 7000 supports enterprise class servicing including non-disruptive code load as well as graceful routing of packets around planned network switch outages.

Unified fabric FCoE, IP, HPC, Ethernet

FCoE is still being defined in standards bodies and won’t be ratified until later this year.  The T11 sub-committee working on FCoE requires 10Gbe and lossless Ethernet switching capability.  As you may recall FC protocol is a truly “loss-less” protocol whereas Ethernet has been designed around a “lossy” protocol providing quick recovery for dropped (lost) packets.  Dropped packets incur added performance overhead and an inherently non-deterministic packet transit time.  By implementing lossless Ethernet today the Nexus 7000 is ready to support FCoE when ratified.

Nexus traffic over FCoE will need to terminate at standard MDS 9500 switches that can translate FCoE to standard FC protocol to talk directly to storage devices.  Servers will ultimately use FCoE directly eliminating the need for FC HBAs.

True unified fabric may need to await FCoE ratification but today Cisco SFS infiniband switches support a unified fabric through infiniband to FC gateways, infiniband to Ethernet gateways and support infiniband HPC switching directly.

10Gbe now, 40-100Gbe ready

The Nexus system is a high density switching architecture with a 32 port 10Gbe and 48 port 10|100|1000 Ethernet blades using 8 blades per unit and two units per system.  Each blade adds switch throughput and in aggregate can support 15Tb/s switch throughput.  Cisco states this high-density switching architecture is 40 and 100Gbe ready.

Server virtualization readily consumes all available port bandwidth and as such is driving the need for higher bandwidth links across the data center.  Cisco has responded by aggressively rolling out 10Gbe.  In contrast, Brocade uses both 10Gbe and 8Gb/s fibre channel to address this rising demand (See SCI StorInt™ Dispatch Brocade080122-010.pdf.)   Both will work and with Microsoft pushing their own server virtualization, the need for more port bandwidth will grow even faster.

Network virtualization

Nexus 7000 supports a virtualized control and data plane that allows for VLANs, Virtual Route Forwarding (VRF), and Virtual device context (VDC).

  • VLANs allow for virtual isolation of bridge domains.
  • VRF provides for virtualized forwarding and routing tables.
  • VDC provides for virtualizing the physical switch itself presenting the switch as multiple logical devices.

Nexus also supports standard storage area network intelligent fabric services using MDS FC switches.

Announcement significance

Cisco’s is making another big bet, this time on the FCoE protocol and 10Gbe to be the unified fabric of the future.  Also Cisco has already rolled out infiniband as another unified fabric.  Alternatively, Brocade has placed bets on 8Gb/s FC and 10Gbe for their unified switching.  Both companies will be watching the market place to see which becomes dominant.

The Nexus 7000 is a major platform change for Cisco, which includes a new branded NX-OS as well as new hardware platforms.  Future versions of Nexus hardware will be targeted to other switching hardware configurations.

A PDF version of this can be found at

Cisco 2008 February 25 announcement of New Nexus 7000 Series data center class switch (PDF 252.6 KiB)

Silverton Consulting, Inc. is a Storage, Strategy & Systems consulting services company, based in the USA offering products and services to the data storage community

 

This Silverton Consulting (SCI) Storage Intelligence (StorInt™) Dispatch provides a summary of Brocade’s recent introduction of their DCX backbone, a new combined 8Gb/s FC – 10Gb Ethernet data center switch.

DCX Summary

Brocade is the first to introduce to market a new 8Gb/s FC and 10 GbE backbone called the DCX.  DCX provides a central core switch for all the multiple, isolated server, SAN, and Ethernet networks in the data center.

DCX is a multi-protocol switch supporting FICON, 8Gb/s FC, GigE, and 10 GbE.  What’s missing from the DCX is Infiniband.  Brocade’s main competitor, Cisco is taking a different approach to network convergence with their Vframe a new data center management framework that uses Infiniband to support FC, GigE and 10 GbE.  The problem with Cisco’s Vframe network convergence is the need for all new Infiniband HBAs, Infiniband gateways,  and a new proprietary management framework.  Brocade believes the DCX, building on current FC and Ethernet hardware, better preserves current data center investments in switch/HBA hardware and current management frameworks while also advancing technology to the next level.

DCX architecture

DCX is configured as a one or two backplane system with power and slots for switch blades.  Brocades FC switch blade supports 48-8Gb/s FC ports.  DCX has an aggregate IO bandwidth of 7.6Tb/s, significantly more than their closest competition.  The two versions of DCX support either up to 384- or 896-8Gb/s FC ports. The larger system consists of two of the smaller systems connected by 4-inter chasis links each consisting of 64-8Gb/s trunked links for an aggregate 1Tb connection between each chasis.  In addition, DCX supports 10 GbE which will ultimately provide support for DCE, FCoE, HPC, iSCSI and FCIP.  Many of these protocols are still being defined in standards committees and will not be available at release but Brocade believes that some may be available as soon as the end of 2008.

One example of an emerging standard is RDMA over Ethernet. Brocade believes RDMA over Ethernet will emerge as a better choice than Infiniband.  Today RDMA, a server-to-server networking protocol is only implemented over Infiniband however a new standard is being defined for RDMA over Ethernet.  Brocade contends when this new standard is adopted the need for Infiniband will be lessened.

The DCX is very power efficient using much less power than other offerings.  In fact, DCX consumes about 0.5 watts per Gb of bandwidth (maximum bandwidth configuration with two power supplies).

In addition, DCX interoperates with current B- and M-series directors and switches.  It is the only product out there that interoperates with both of these products which together own the majority of FC ports in the enterprise.

Brocade products in this class provide high reliability, non-disruptive blade and technology upgrades, allowing the DCX to be your data center’s network backbone for a long time to come.  DCX is protocol agnostic and as new protocols emerge they can be readily adopted.  In our discussions on the lack of Infiniband Brocade said they were watching the technology and if it becomes successful could readily implement an Infiniband blade for DCX.

Regarding the market place, some believed the need for 4Gb/s FC was never going to materialize but the rapid adoption of server virtualization put that discussion to rest.  Brocade is starting to see some customers supporting 40-50 virtual machines per physical server, causing the need for even higher bandwidth.  Thus, the need for 8Gb/s is evident today and as server virtualization continues will only become more acute.

Announcement significance

First and foremost the delivery of an 8Gb/s FC backbone is significant.  The addition of 10GbE to the FC switch backbone makes a lot of sense if the standards can be approved to support the more QOS protocols like RDMA over Ethernet, FCoE and DCE over 10GbE.

Network infrastructure changeover is a multi-faceted process involving a delicate step-by-step progression of switch, HBA, server, and storage vendors.  With this announcement at least one switch vendor is ready to support 8Gb/s FC.  It’s now up to rest of the industry to follow suit.

A PDF version of this can be found at

Brocade 2008 January 22 announcement of Brocade's new DCX Backbone (PDF 228.4 KiB)

Silverton Consulting, Inc. is a Storage, Strategy & Systems consulting services company, based in the USA offering products and services to the data storage community

For a no risk, free 30 minute consultation please give us a call at +1-720-221-7270

If you are interested in receiving our free monthly newsletter on major storage announcements and storage performance results please subscribe by email
Silverton Consulting, Inc., P.O. Box 492
Broomfield, CO, 80038-0492, USA
© 2004-2012 Silverton Consulting, Inc., All Rights Reserved
Contact|SCI Privacy Policy|Sitemap
Suffusion theme by Sayontan Sinha