Every business-critical application requires a reliable path to its data. That data needs to be available, and that data needs to be protected. Today, we find ourselves challenged as the world throws seemingly continuous disruptions at the very IT organizations tasked with keeping every enterprise’s data safe.
Whether a pandemic, a fire, hurricane, or even an errant backhoe, external forces all seem to conspire with bad actors who spend their nights keeping your cybersecurity teams awake. Any one of these can disrupt even the most resilient and well-thought-out of plans.
IT organizations that embrace fast and flexible data center architectures are the ones who find themselves much better poised to face these disruptions. The organizations that deploy resilient enterprise storage solutions based on reliable and flexible data paths are the organizations that are best poised to deliver data to wherever it’s needed the most. Data, after all, is the foundation of everything that IT does.
I had the opportunity late last year to talk to a group of experts from Broadcom’s Brocade storage networking group and IBM’s storage division. Broadcom had just launched its new Fibre Channel Gen 7 portfolio, a collection of switches, hubs, and HBAs that double the previous generation Fibre Channel devices’ performance. IBM had just adopted the Gen 7 technologies in its products. Gen 7 delivers up to 64Gbit/second. Gen 7 sets the stage for fast Fibre Channel and the coming generation of NVMe-over-Fibre-Channel solutions.
During the conversation, I learned that Broadcom’s Brocade storage networking fabric, when coupled with IBM’s intelligent storage technologies, provides a rock-solid foundation for moving data throughout an enterprise. It was an eye-opening conversation, touching on Fiber Channel’s evolving abilities to provide a self-healing, self-tuning, high-performance, and resilient storage infrastructure.
Cyber-resilient storage architecture
Cyber-resilient architectures define an organization’s ability to weather the storm. If we look at just the past year, we’ve seen a global pandemic, a historic hurricane season, earthquakes, forest fires, and an unprecedented number of debilitating cyber-attacks. While it can feel like it’s all landed at once, the reality is that these sorts of events happen continuously. And they disrupt everyone’s plans, especially IT’s.
AJ Casamento, a Broadcom Solutioneer, explains three key elements to think about when building a cyber-resilient storage architecture: data security, data protection, and data recovery. Adjusting those three elements to address your recovery point and recovery time objects will directly impact how your storage system is architected.
The foundation of any storage solution is the storage array. AJ pointed out that, for example, IBM’s latest generation mainframe storage, the IBM DS8900F, directly addresses cyber-security with its built-in end-to-end encryption. This even includes encrypted paths to virtual tape subsystems. Putting technology like IBM’s Fibre Channel Endpoint Security on top of a Brocade storage fabric, with its data compression and encryption capabilities, allows for secure and efficient transport of data without any fabric congestion or overall performance impact.
This is true whether or not you’re using IBM’s embedded security. As AJ says, “if you don’t have encrypted data, we can encrypt it for you, delivering a hybrid model of old and new, allowing you to migrate into a fully encrypted environment while still having your same infrastructure in place.” That’s powerful.
IBM, for its part, delivers a robust set of cyber-security capabilities throughout its line. The mainframe-focused DS8000-series doesn’t just provide end-to-end hardware-accelerated encryption, but it supports immutable, safe-guarded copies of protected data allowing for rapid restoration. IBM’s Brian Sherman pointed out that IBM’s Spectrum Protect portfolio allows quick detection of anomalies in backup and snapshot data.
An autonomous infrastructure
One of the most impactful trends of the past decade has been the adoption of autonomous systems. Systems are becoming self-learning, self-optimizing, self-tuning, and even self-healing. IBM and Broadcom sit front and center in this world. As AJ Casamento points out, systems are getting too fast for humans to be in the critical path.
Self-learning infrastructure means that the storage fabric learns over time what normal behavior looks like to rapidly highlight when there’s a deviation from normal. This allows the infrastructure to raise alerts to IT teams and, in some cases, take corrective action independently.
The fabric’s ability to take corrective action is the defining core of a self-healing network. When there is congestion in the fabric, a switch needs to rapidly isolate failures and open alternate paths for congested data flow. Let the humans look at it later but keep those business-critical applications running. It’s application availability, after all, that matters most to business owners.
Broadcom’s Brian Sherman highlights the technologies that underpin the ability for this level of self-learning and self-healing, including Broadcom’s Brocade SANnav Management Portal and SANnav Global View. The SANnav products, when coupled with ASICs embedded within a Brocade fabric, go beyond simple SAN management and allow for deep-packet inspection to occur at line speed.
Marrying the Brocade SANnav products with IBM’s Storage Insights capabilities yields a compelling solution. It’s one that provides management capabilities that range from the routine to the AI-driven self-learning, self-healing world that we’re all rapidly moving into.
Fibre Channel isn’t going anywhere
It’s easy to get wrapped up in the idea that old technologies should be rapidly replaced by the latest and greatest. I know that I’m consistently guilty of that thinking. The reality is, when you’re protecting your organization’s data, you want an infrastructure that is both rock-solid and highly-performant. New technologies, such as NVMe-over-TCP, have a place, but bedrock mission-critical data still depends on evolutionary and stable technologies.
Fibre Channel has been around for a very long time. There is likely no big data center or enterprise in the world that doesn’t have an installed fabric. It just works and, with a performance jump like that provided by Gen 7, it continues to get better.
NVMe-over-fabric is the future of storage transports. It’s a technology that received a considerable boost when VMware announced support with its release of vSphere 7 last fall. NVMe-over-fabric is a capability that can be provided by high-speed ethernet, Inifiband, or Fibre Channel. Gen 7 Fibre Channel delivers the speeds and latencies required to take advantage of NVMe-over-fabric fully. 64Gbits a second is fast.
I love the idea of NVMe-over-IP, but it is still very early days for those solutions. The players in that world are still figuring out how to deliver the rock-solid performance required to be a storage fabric. It’s a mixed bag of RoCE-enabled ethernet adapters and a ream of tuning suggestions. The ability to scale NVMe over ethernet is still very much unproven. Anyone who has ever managed an iSCSI network trembles just thinking about it.
Running NVMe over Fibre Channel allows an enterprise to leverage its existing storage fabric investment. IT shops can use the same tools they’ve long used to manage fabrics, evolving to support the new protocols. The beautiful thing about running NVMe over Fibre Channel is that we know that it will just work. Nobody should be declaring Fibre Channel dead.
I enjoyed the conversation with AJ Casamento, Brian Sherman, and Brian Larsen. That group has worked together for a long time and, like all veteran technologists, had great war stories to tell. I encourage you to listen to the podcast or watch the video.
The future of storage fabric is NVMe. With its latest Gen 7 capabilities, Fiber Channel gives enterprises the ability to migrate to NVMe-over-fabric architectures with confidence quickly. Confidence that it will just work, that there won’t be a lot of re-training that has to happen, and that long-term investments in its storage infrastructure continue to pay-off.
IBM continues to deliver storage solutions that take advantage of these new capabilities. The relationship between Broadcom and IBM is a deep one. I’ll give AJ Casamento the last word.
“The relationship between Broadcom’s Brocade and IBM is well into 30+ years. The confidence of what we’re doing together is better together because we test our solutions together. We look at advanced technologies and make those real-world so that people will realize that it’s ready for prime-time.” They’ve convinced me.
Note: Moor Insights & Strategy writers and editors may have contributed to this article.
Moor Insights & Strategy, like all research and analyst firms, provides or has provided paid research, analysis, advising, or consulting to many high-tech companies in the industry, including 8×8, Advanced Micro Devices, Amazon, Applied Micro, ARM, Aruba Networks, AT&T, AWS, A-10 Strategies, Bitfusion, Blaize, Box, Broadcom, Calix, Cisco Systems, Clear Software, Cloudera, Clumio, Cognitive Systems, CompuCom, Dell, Dell EMC, Dell Technologies, Diablo Technologies, Digital Optics, Dreamchain, Echelon, Ericsson, Extreme Networks, Flex, Foxconn, Frame (now VMware), Fujitsu, Gen Z Consortium, Glue Networks, GlobalFoundries, Google (Nest-Revolve), Google Cloud, HP Inc., Hewlett Packard Enterprise, Honeywell, Huawei Technologies, IBM, Ion VR, Inseego, Infosys, Intel, Interdigital, Jabil Circuit, Konica Minolta, Lattice Semiconductor, Lenovo, Linux Foundation, MapBox, Marvell, Mavenir, Marseille Inc, Mayfair Equity, Meraki (Cisco), Mesophere, Microsoft, Mojo Networks, National Instruments, NetApp, Nightwatch, NOKIA (Alcatel-Lucent), Nortek, Novumind, NVIDIA, Nuvia, ON Semiconductor, ONUG, OpenStack Foundation, Oracle, Poly, Panasas, Peraso, Pexip, Pixelworks, Plume Design, Poly, Portworx, Pure Storage, Qualcomm, Rackspace, Rambus, Rayvolt E-Bikes, Red Hat, Residio, Samsung Electronics, SAP, SAS, Scale Computing, Schneider Electric, Silver Peak, SONY, Springpath, Spirent, Splunk, Sprint, Stratus Technologies, Symantec, Synaptics, Syniverse, Synopsys, Tanium, TE Connectivity, TensTorrent, Tobii Technology, T-Mobile, Twitter, Unity Technologies, UiPath, Verizon Communications, Vidyo, VMware, Wave Computing, Wellsmith, Xilinx, Zebra, Zededa, and Zoho which may be cited in blogs and research.