Jayshree Ullal the former star Cisco Senior Vice President responsible for Cisco's $10 billion in revenue Datacenter Nexus 7000, Catalyst 4500 and 6500 switch product group and services line, has embraced cloud networking as her next gig by joining high-performance switch vendor Arista Networks as its President and Chief Executive Officer.
"I am delighted to join Arista.
"The combination of an outstanding team, innovative hardware and software technology and the Cloud Networking vision drew me to this company. "I look forward to creating a unique focus for our strategic customers, partners and channels," said Ullal. |
Obviously excited about her "next gig" as she calls it, Jayshree detailed Arista's Cloud Networking vision in her blog yesterday.
"Cloud networks simply cannot be achieved with today’s monolithic networking software stacks. I am very excited by Arista’s Extensible Operating System (EOS) innovations and its impact on Cloud Networking.
"By separating process and state, EOS can distribute software and configuration information across the entire fabric, enabling seamless consistency across the infrastructure. This in turn enables to provision clouds, add new servers or services, and update images and patches with simplicity and ease, and to automate every step of the provisioning process."
Jayshree adds that there are many attributes that customers need to consider for cloud network fabrics that directly affect the choice of switch and software architecture. Some of the key considerations are below (the text and graphics below are drawn from Jayshree's blog as well as other information found on Arista's Web site):
| 1. Scalability:
The cloud network must scale to the overall level of throughput required to ensure that it does not become the bottleneck. This means the cloud networking fabric must handle throughputs that will reach trillions of packets in the near future. |
2. Low Latency:
The cloud network must deliver microsecond latency across the entire network fabric since low latency improves application performance and server utilization. For latency sensitive applications, 10 Gigabit Ethernet is a major improvement: |
3. Guaranteed Performance:
The cloud network must provide predictable performance to service a large number of simultaneous applications in the network, including video, voice and web traffic. EOS: An Extensible Operating System The key to EOS benefits is its unique multi-process state-sharing architecture that consists of multiple processes interacting with a central shared state repository called Sysdb (system database). This architecture is shown below. EOS Multi-Process State-Sharing Architecture EOS derives its benefits from the essential characteristics of the architecture above:
|
4. Extensible Management:
Real-time upgrades and image/patch management in a large cloud-network is a daunting challenge to network administrators. A vastly simpler approach is required to handle networks of this size, which automates provisioning, monitoring, maintenance, upgrading, and troubleshooting. The EOS multi-process state-sharing architecture is also the key to reducing maintenance windows by allowing more maintenance tasks to be performed during normal switch operation. EOS’ in-service software upgrade (ISSU) capability relies on the ability to restart the patched agent without disrupting switch operation. Packet forwarding is unaffected by the module upgrade process, so there is no user-perceptible downtime. ISSU is illustrated below. ISSU - EOS’ in-service software upgrade |
5. Self-Healing Resilience:
Cloud networks operate 24x7, so downtime is not an option. This requires a network architecture that offers self-healing and the ability for transparent in-service software updates. On most switches, any software fault results in a reload, resulting in seconds or even minutes of downtime. Under EOS, any fault is contained within the agent or driver where the fault originated. If the fault causes the agent to crash, then the EOS process manager (ProcMgr) restarts it immediately. If the fault causes the agent to hang or loop, ProcMgr detects the condition and restarts the agent. Thus, faults within EOS are self healing. This process is shown below. Fault Containment and Self Healing |
Related stories:
Ex-Cisco exec drawn to start-up Arista's software architecture
Ex-Cisco exec Ullal lands at data center start-up
Former Cisco rising star to head up data center switching firm
Do you think Jayshree has made the right choice embracing cloud networking as her next gig?
One year warranty on refurbished:
One year warranty Cisco Repair
One year warranty Refurbished Cisco
|
Cisco vs. Competitor Lab Tests |
Brad Reese cofounded BradReese.Com Cisco Refurbished, which enables affordable Cisco networks globally by assuring customer satisfaction with guaranteed one year warranties on both Cisco Repair as well as Refurbished Cisco.
Don't be shy, contact Brad Reese online or call him at 646-827-1130.