Not quite a year ago, Cisco and Google announced a Cloud partnership. Today, at the very first keynote at Cisco Live 2018, Diane Greene, CEO of Google Cloud joined Chuck Robbins on stage to talk about the partnership, highlighting Kubernetes and a  unified security policy.   Both Chuck and Diane want a large ecosystem of partners and developers.  Later on, Chuck mentioned Cisco passing the 500K developer milestone for DEVNET.

Chuck touched a litter on routing, mentioning next-generation branch and highlighting intent based networking activity in the SP space.  For example, one of their SP customers updates 60,000 routers each night using automation. He then quickly got back to the Catalyst 9K switch, highlighted as the fastest ramping product ever in Cisco history.
 
Children’s Hospital of Los Angeles was highlighted as a customer of the Cat 9K. As a customer, they have over 35,000 connected devices.  They purchased over 2,000 WLAN APs and over 200 Cat 9300 switches. They are also deploying ISE and have done over 23,000 different device profiles/identifications and are track to start policy enforcement.  They are now in the process of deploying at their branch locations. They noted 550k blocked threats over the first few months of deployment.
 
My key takeaways are that there is an explosion of devices and data on the network, much of which is encrypted and a human can only do so much; thus the network must scale and automate.  Cisco is looking to use AI, automation, and its architecture to allow the customer to scale with those IoT devices and to have the network automate many tasks, especially around security. Monetization for Cisco will occur both in the hardware, but also in the solution sale.  An ideal customer would be end-to-end Cisco, but Cisco will also support open APIs in order to allow partners and customers to operate with their preferred solutions.

Last week we attended Huawei Connect conference in Shanghai which is turning into a massive event for Huawei with significant customer attendance.  It was tons of fun to talk to customers, catch up with friends, and the different parts of Huawei while on the show floor.  While there are ton of highlights from the show, here are a few highlights that peaked our interest.

• Huawei became the first vendor to offer an 802.11ax Access Point (AP).  Sampling now and shipping in early 2018.  We ran into quite a few engineers on the show floor that were very proud of being the first vendor to announce a product in this space and learned quite a bit beyond the spec sheet.
• Huawei customers are broad based and growing significantly.  We ran into many large customers, especially from western Europe, that are looking to partner with Huawei beyond just point product purchases.  In some cases, this was campus switching and WLAN, in others it was DC switching and campus.  Huawei is doing a very good job of providing solutions and not just point products and customers are embracing it.
• 1% of customer spend.  Huawei consistently highlighted that it was only targeting 1% of customer spend, leaving the other 99% to its partners.  We found this interesting that as Huawei grows they are trying to assure their partner community that they will not go after the entire software and solution stack and instead focus on the pipes and compute.  In areas of IoT they would integrate deeply with those partners, but the goal was to provide a complete selection directly to the customer.

There were a ton of other highlights, we are happy to discuss, if anything sparks your interest, let us know.

Today Nokia announced its new FP4 ASIC and 7750 SR Router.  Playing the leapfrogging game on speeds, we saw 36  400 Gbs ports in a 2RU box that looks awfully similar to a spine switch and the further blurring of what a next gen router and switch really look like, especially in the Cloud.  

We heard continued confusion over winning Cloud scale accounts.  We note that a customer like Apple buys from multiple vendors and for multiple reasons.  What Apple builds for their own consumption is not what they will deploy in a telco provider or peering location.  

The debate between merchant silicon and custom ASICs continues to come up.  While we are slightly in favor of merchant silicon, we note that the Cloud providers do not fear custom ASICs, they merely want to have standard APIs to control that equipment.

We note the Nokia ports are DDQSFP and not OSFP so we do not have a clear answer on form factor either.  We now wait for the next product announcement with the only clear answer that we are in a phase of rapid innovation in order to keep up with the network traffic demands of the Cloud.

Today Broadcom announced Trident 3.  The companies third major release of a chip that drove the merchant silicon revolution in the data center and started the white box movie in the Cloud.  With Trident 3, all of Broadcom’s data center switching ASICs now support speeds of at least 3.2 Tbps per chip.
 
Trident 3 is impressive, but a few things about it really caught my eye.  First, Trident 3 will offer five different skus, two of which are really focused on campus switching.  One could see a 48-port 2.5 Gbps switch out of the X3 version of Trident 3 next year.  We believe the Trident family moving into the campus will be significant for the industry once products begin to ship.
 
Second, native 25 Gbps ports.  Trident is the most popular of Broadcom’s ASICs, especially in the enterprise, and with Trident 3, we expect the market to quickly move away from 10 Gbps/40 Gbps products and towards 25 Gbps/100 Gbps products.  This product aligns well with our forecasts for this transition which we are excited to be publishing shortly.  We still don’t see a bandwidth need in most enterprises for 25 Gbps, but the ability to future proof at the customer level and the ability to consolidate skus at a vendor level will make this compelling.
 
Third, we see a potential for both switch vendors and customers to benefit from one family of ASICs from the campus all the way to the data center.  While it is too early to know the impact of this right after the announcement, we look forward to conducting interviews over the next few months to define this impact.

​It was great to catch up with old friends and make new friends at OCP this year.  The show was highly successful with attendance at the Facebook and Microsoft booths so large that it was difficult to move around.  On the switch side, most of the announcements were incremental to the market, but with new chips on the horizon, and a delay in 100 Gbps because of supply constraints, we see this as a temporary pause ahead of what will likely be some bigger announcements in 2018.

There were many highlights at OCP, but three things caught our eye while walking the show floor on both days.

•    Microsoft’s project Olympus server is about to transition Microsoft away from High-Density servers and towards Rack servers.  This is more in line with what other Tier 1 cloud providers are doing.  We note the smart-NIC is still a multichip solution, one that could be further reduced in future generations.  They also announced ARM based servers and joined Facebook on announcements in machine learning and AI optimized compute.  We see this change in Cloud architectures as a good sign for the industry.  The market is quickly moving into more use cases, which will help drive growth beyond just moving workloads away from the premise market. 

•    The white box vendors were in force at the show.  Edgecore showed various Fixed and Modular form factors.  We note that some of these boxes are modified for larger Cloud customers with the inclusion of large SSDs or memory.  We have a pretty good sense of what is using these additions, but that is a topic for a more detailed report.  We also saw Quanta and Delta with large presences on the show floor.

•    This year we saw many software announcements around OCP.  Arista announced their containerized EOS operating system (cEOS).  We saw Apstra and Cumulus active at the show as well running into many other software vendors in attendance.  OCP has done a good job at straddling the hardware/software boundary, but clearly the software needed to run these networks is quickly evolving as well.