Apple’s Cloud business remains locked between business models.  Apple monetizes the Cloud portion via hardware sales with little direct revenue coming from Cloud services. 

While Apple announced a significant networking advancement with Nokia’s new data center switches, a substantial step in the right direction, we have yet to see a significant increase in organic data center spend. 


Instead, Apple utilized other large IaaS providers for a lot of Apple’s features, which, to a certain extent, we see a conflicted to Apple’s vertical integration of the company's portfolio.  We expect Apple’s organic data center spending to increase significantly over the next few years as the company further vertically integrates and competes directly with Google/Facebook on features and capabilities even though the three companies monetize them differently.   For example, computing is just as critical in Augmented Reality and Virtual Reality as the handset itself.

The only negative in Microsoft’s earnings in 2Q20 was Bing, expected based on consumer behavior trends.  Azure continues to grow robustly as the company experiences supply constraints, just like everyone else in 2Q20.  Over the next 4-6 quarters, we expect Project Jedi, the DoD’s Cloud project awarded in 2019 to Microsoft, to ramp up, which will be a significant positive to demand during the quarters in which it recognizes revenue. 

Microsoft is not only benefiting from enterprises using Azure for application development but also in collaboration with Teams and Office 365.  Almost all WFH scenarios help Microsoft, except for higher unemployment (less total office workers).

Microsoft remains the dominant #2 IaaS provider and is increasing its influence in the supply chain on future projects and deployments. In 3Q20, we expect LinkedIn revenue to moderate based on fewer companies hiring.  We also expect the new XBox, Series X, to be a key driver for Microsoft’s Cloud services.

Facebook revenue grew robustly in 2Q20, and its CAPEX guidance remained consistent for 2020 compared to previous revisions in the last two quarterly results.

Facebook’s results were counter trend to our expectations that many advertisers would pull back spending do to COVID-19 based on lack of supply (no need to advertise consumer staples) or from consumer spending put on pause (cars that people don’t need while sheltering at home).  We believe Facebook benefited from more time on the platform and from the targeting of specific adds on areas of discretionary spending that did grow like sports equipment (good luck finding a bike, kayak, or other social distancing sports gear) and WFH (consumers shifting patterns of spending more for their residents while WFH or just making the home more comfortable due to extended hours in it).

Facebook, like Google, is under government scrutiny for its scale and size.  We are closely monitoring the trends of government oversight from the US government and well as other countries like Australia which is forcing Facebook to pay for news as well as Microsoft’s potential purchase of Tik Tok (as of writing this over the weekend, they were still pursuing them).  It seems like the duopoly here is not preferred by most governments at this stage and expect election results to polarize the losing party against social media companies into 2021.

Google, the largest US Hyperscaler by revenue, reported Search and Social results that declined Y/Y for the first time while IaaS revenue grew nearly $1B Y/Y.  We were a little surprised at Facebook’s robust growth compared to Google’s. Google’s results were in line with our overall expectations for Search and Social decline in 2020 as consumers and advertisers resetting to the new normal.  We expect more targeted ads throughout 2020 as consumers live and work from home, and many students live and study from home during the fall semester.

Google has made big bets and investments in IaaS, and we continue to see AI as an area where they will attack AWS and Azure.  It is unclear if IaaS is compatible with the culture withing Google, which could put an upper limit on the verticals and companies Google can sell to.  During 1H20, Google was surpassed by Amazon in our supply chain interviews as the company with the most influence on the technological direction of industry-wide future products. 

We see a passing of the guard as AWS CAPEX is now much higher than Google’s, and the supply chain sees Amazon as more significant revenue potential.  We expect this change to reverberate throughout the supply chain, primarily based on how each Cloud provider uses custom or semi-custom semiconductors in their data center infrastructure.  This is something we are happy to talk about as we prepare our 2Q20 results and our fall readouts.
-- Alan Weckel, Founding Analyst, 650 Group

Over the next five days, we will highlight each of the US Hyperscalers and the results they had during 1H20 and 2Q20. Today we will start with the overall trends in the market. US Hyperscaler revenue grew 20% in 2Q20 compared to a year ago, setting a new record.

US trade war activities, mainly against Huawei, caused significant lead time increases in many critical components for Cloud data center build-outs during the quarter as the 5G battle against China is having ripple effects into the Cloud supply chain. Custom ASIC and semi-custom ASIC development in the Cloud continue to expand with multiple new initiates around #AI #ML., #SmartNICs, accelerators, and #CPUs underway. This is not to mention Amazon Web Services (AWS) getting into #6G with a $10B investment in project #Kuiper for low earth satellites in direct competition with @SpaceX Starlink. There are over 50 custom ASIC projects in the Cloud. Each one has implications on the supply chain and immediate potential to shift market share from each Cloud provider.

Our overall projections for data center spend in switching, servers, and storage remain relatively unchanged since our previous forecast. Current results affirm our forecasts as we shift to vendor performance over the next two weeks, which we expect to be dependent on each company’s vertical and enterprise exposure.

Before the consolidation in the Mobile Radio Access Network (MRAN) market that occurred in the past decade during Huawei’s ascendancy, there were a dozen major RAN vendors.  They included Motorola, Lucent, Alcatel, Siemens, Nokia, Ericsson, NEC, Fujitsu, Samsung, Nortel, Huawei and ZTE, and they hailed from the US, France, Germany, Finland, Sweden, Japan, Korea, Canada and China. 


As Huawei entered the market, using a price aggressor strategy, it catalyzed mergers, resulting in the elimination of Motorola, Lucent, Alcatel, Siemens, Nortel, plus a reinforcement that led to the Japanese and Korean players to sell primarily to their home markets.  The result is that in many markets during recent years, there were only two vendors left, and that left operators with little choice but to look elsewhere. 

The punchline is that going forward, due in part to Open RAN, and in part to the response of operators looking outside their traditional supplier base, we now have 10 RAN players who can bid on projects.  And there is a multiplier on top of the 10 players, because going forward,  operators can buy radio heads from different vendors than their primary RAN baseband vendor, essentially doubling the number of choices an operator has when making mobile RAN vendor decisions.

Here is how we arrive at the conclusion that there were only two players per major geography.  Just a couple years ago, the state of affairs was quite different; we had only Nokia, Ericsson, Huawei and ZTE as players, and last year, it became clear that in the US, only two major players were left.  In China, the same could be said, with Huawei and ZTE as main suppliers (Ericsson has won business there and Nokia ceded the market in 2019).  And in 2020, we’ve seen much of Europe and English-speaking Asia whittle down to two suppliers, as well.

And here is how the procurement teams at operators have much choice in the future.  The “Open RAN” vendors are now deemed viable given the success at Rakuten and the push by operators to demand Open RAN compliance, and these include Altiostar, Mavenir and Parallel Wireless.  Nokia and Ericsson are invited to most, if not all bids worldwide. 

Huawei and ZTE are invited to many, but a declining number of bids in markets that are siding with the US viewpoint.  We saw a turning point in late 2018 when AT&T announced it will buy from Samsung, who has now gotten a strong foothold in both India and the US.  And, more recently, we have seen two Japanese players, NEC and Fujitsu, in some way filling in the void left by Huawei and ZTE’s woes in the US/China spat, as they get wins (Fujitsu recently won DISH) and get invited to bid (NEC and Fujitsu are being asked to bid on UK projects).  Add these up and we have Altiostar, Mavenir, Parallel Wireless, Nokia, Ericsson, Huawei, ZTE, Samsung, NEC, Fujitsu. 

There are other factors at work that are adding to more RAN choices, as well.  Two such trends are Facebook’s efforts, ONF’s efforts and the variety of radio head vendors who are now viable with Open RAN/FB/ONF efforts.  Facebook has promoted projects such as Telecom Infra Project (TIP) that have many goals, including one that supports the goal of $1,000 radio heads (these cost much more from the major vendors). 

The Open Network Foundation (ONF) supports projects such as SD-RAN and Aether. 
Radios can be purchased from non-traditional sources, as well because with all three projects we have mentioned above (TIP, ONF SD-RAN and Open RAN), these allow radio purchases to be made separately from baseband purchases, literally doubling the choices that operators have when building out a roster of vendors.

The trends in mobile RAN have changed significantly.  Vendors with little to lose (startups and players entering new markets) are getting aggressive to grow their businesses.  Incumbent vendors are at risk, as their business practice of selling baseband and radio simultaneously to captive operators is coming to an end.  We may look back at this early 5G era and say there was a lot more to it than just the upgrade to 5G, and it begs the question, who will acquire whom to consolidate the market once again and get pricing under control.

We attended two separate presentations made by Ciena last week and have reflected on the comments made by the company.  In summary, Ciena advocated using both pluggables like ZR and ZR+  as well as high-performance optical transport systems (its main business) together to construct cost-effective networks.  The mix and match recommendation serves Ciena well, in that, substantially all of its revenues are high-performance systems, and coherent pluggables are a substitute threat to its business.  If its customer base wanted to adopt pluggables but continued buying systems from Ciena, it would be logical for the customers to consider both systems and pluggables.  Ciena argued that its pluggables would be superior to competitors, as well, highlighting its unique DSP, PIC and packaging as best in class.  We find the pitch could be compelling if Ciena’s pluggables are better, and would play to Ciena’s advantages.

Much of Ciena’s recent growth has come from cloud hyperscalers.  Hyperscalers currently use Ciena’s systems equipment for the Data Center Interconnect (DCI) use-case - to connect one data center to another.  We have forecasted that hyperscaler DCI networks will move rapidly towards coherent pluggables, once available, substituting for high-performance systems.  In Ciena’s presentation, it agreed with our assessment that short-haul DCI is the first place where pluggables will be put to work, displacing optical transport systems.  Our view is that, at the market level, the metro optical transport systems' revenue path in 2021 and beyond will decline based on the transition to pluggables-use by DCI networks.  Ciena is wisely hedging its bets by offering both pluggables and systems.  But, we don’t think pluggables-related revenues will offset the potential loss of systems revenue, especially if the move towards pluggables is fast.  One thing that Ciena has in its favor during this transition is it took first revenues on its 800 G class of systems equipment in its April 2020 fiscal quarter; its early 2020 launch could put Ciena back in the driver's seat again with customers who demand very high-performance optical links.  With the inclusion of 800 G systems, Ciena’s systems offerings will be more competitive than it was entering 2020, and more competitive than 400 G class of pluggables that are the primary topic of this article.  At present, only Ciena’s competitor, Infinera, has planned a 2020 launch of 800 G class of equipment.

Here are some specifics from Ciena’s two presentations (alternative link) last week. Ciena’s view is that single Span Data Center Interconnect (DCI) and High Capacity Access (Metro) are the most likely markets to adopt 400 Gbps ZR or ZR+ optical modules.  It says that multi-span metro may have some use for ZR/ZR+, and long-haul and subsea won’t leverage these pluggables in the near to medium term.  We agree with this assessment.  Ciena will be offering coherent pluggables in two ways, as part of its packet networking and optical systems portfolio, as well as offering them through its Microsystems business for use in 3rd party equipment.  

Ciena shared its assessment of its capacity versus reach comparison of its pluggables (56 GBd) and its high-performance systems (95 GBd).  In the capacity graph the company shared during its presentations, you can see that Ciena’s tests show that coherent pluggables generally have half the reach or half the speed.  The company advocates for “mixing bauds,” which means that for networks that are more complex than simple point to point DCI networks – for instance, ones that have lots of ROADMs, it makes sense to use high-performance metro DWDM systems as well as switches/routers with coherent pluggables.  By “mixing bauds,” Ciena says it expects to 100% coverage of complex metro/regional networks (typical of telcos).

The company explained that it uses four major components in making 400 G coherent pluggables:

• CMOS Technology node.  7nm finFET (35% the size of 28nm planar CMOS) 
• SiPhot PIC.  
• Advanced Packaging Technology.  Multi-die packaging.
• DSP.  Multiple FEC options.  Embedded CPU.  Client framing

Alan Weckel:  This week, HPE Aruba announced its planned acquisition of Silver Peak for $925 million with an expected close date during HPE's fiscal 2020, so we expect the deal to close in calendar 4Q20. SD-WAN is becoming a battleground for vendors as an increasing portion of enterprises want single-pane and cloud-managed solutions for their branch equipment.  

Our research indicates the average branch employee connects to over one dozen cloud services daily and the edge-to-cloud experience has never been more critical in the COVID and post-COVID world as businesses try to engage with customers across an increasing amount of physical and virtual locations.  

Enterprises will accelerate the retirement of MPLS links for more advanced WAN architectures with Cloud-first principles in their physical branches.  Enterprise will also deploy a mix of hardware and software into employee's houses as Work-From-Home (WFH) changes every employee's residence into a new branch extension of the enterprise.  

We have conducted significant research into the Branch in 2020 and how it will transform across verticals and product categories (SD-WAN, legacy routing, WLAN, switching, and security) based on long-term structural changes to the market as well as the sudden changes forced on customers by COVID-19.  We project SD-WAN as the fastest-growing component in the Branch over the next five years.
Silver Peak's advanced SD-WAN portfolio (both hardware and software), completes HPE Aruba’s breadth of WAN offerings, strengthens Aruba ESP (Edge Service Platform) and complements Aruba's strong position in Ethernet Switching, and WLAN, especially Aruba's newer WiFi 6 APs and custom ASIC-based 6200/6300 Access switches.

We expect that AI/ML will increasingly become part of the Branch market in each year of our forecast driven by further product integration by vendors and the need to control and monitor an increasing amount of user and device (IoT) traffic.  As the AI engines learn, self directly networks will move towards self-driving as automation of tasks beyond simple device management become common across networks.  The ability to use AI at this scale in networking today is limited to the largest hyperscalers, but will quickly make its way to campus and branch networks.

We expect overall Branch spending to increase above its 2019 run rate during the forecast period (post-COVID) and single-pane management to nearly triple during that timeframe.  HPE Aruba's acquisition of Silver Peak will help them address the fastest-growing part of SD-WAN and Branch networking.  While there are dozens of SD-WAN, campus switching, WLAN, and security vendors vying for Branch spend, there are only 6 US-based vendors that have a holistic portfolio that customers want and need.

Posted by Alan Weckel, founding technology analyst.

Today, Nokia announced the company's entry into the Data Center Switching market with new hardware and an open Linux-based network operating system built from the ground-up to address the needs of hyperscalers.  Similar to a startup approach, Nokia started in stealth and with an open whiteboard to create a brand new operating system, automation toolkit, and set of switches to address hyperscalers' needs going forward.

Today's hyperscalers are very different from just 3-4 years ago in terms of scale and sophistication.  Our US Top 5 Hyperscaler segment includes Amazon, Apple, Facebook, Google, and Microsoft, and each of them purchases at a scale never seen before in networking.  To put this in perspective, the average Hyperscaler buys more networking gear in a month than the entire installed base of the largest Telco SPs and enterprises.  

The scale at which hyperscalers operate is very different and new for the entire supply chain to adjust.  Automation, AI, and agility become critical as humans cannot scale manual tasks in the same way.  Operational tasks, from configuration to maintenance, must be scaled with minimal human involvement. Nokia's network operating system embraces the approach of modern built-for-cloud architecture from its Linux base, allowing it to run on multiple ASICs as needed. Its operation and automation toolkit includes a "NetOps Development Kit" to enable easy integration of operator network applications and tools, and a "Digital Sandbox" that provides a unique ability to test and roll out new applications and versions of its operating system without risk.  At cloud-scale, there is no longer a dedicated dev/test/production network, so the concept of testing before running in production is very different from the typical enterprise application workflow.

While Nokia's new products were purpose-built for hyperscalers, other segments of data center switching will also benefit from the flexibility and openness of SR Linux, the automation tools, and high-perforamnce hardware. Large enterprises, colocation providers, tier two cloud providers, and telco service providers increasingly follow the architectures and design principles of hyperscalers as they work through their digital transformation projects.  

Hyperscaler networking tiers continue to increase based on several factors.  In the leaf/spine (Top-of-Rack/Aggregation) layer of the network, Hyperscalers are adding levels because of increased utilization of the network and the server.  Increased utilization is coming from the use of Smart NICs capable of offloaded certain functions like security, storage, and virtualization, allowing the server to run more efficiently at a higher percentage of the time.  Improved utilization creates an increased network load that changes the leaf/spine portion of the network.  Hyperscalers also deployed the virtues of SDN and AI to use telemetry data to understand better and test the network, allowing it to run at higher utilization.  The above trends will rapidly move the hyperscalers towards 100 Gbps server access and 400 Gbps aggregation networks.

In the Core, Ethernet Switching and Routing is rapidly adding additional tiers to address the needs of inter data center connectivity.  Data Center Interconnect (DCI) is about to become a new multi-billion dollar market for Ethernet Switching and Routing.

On the hardware front, Nokia announced Fixed and Modular platforms based on Broadcom's Jericho L3+ ASICs and Tomahawk L2 ASICs and the ability to adapt to future ASICs based on Cloud demand.  By filling out the company's portfolio, Nokia has become one of just a few companies that can address the Cloud networking needs from server access all the way to transport.  Given the increasing demands of multi-vendor and multi-ASIC from the hyperscalers, the new vendor diversity will help drive the ecosystem as a whole at 400 Gbps and beyond towards 800 Gbps and 1.6 Tbps based platforms in the future.  

At Cisco Live, Cisco announced its next Cloud Scale ASIC.  Specifically, it announced it's 25.6 Tbps (512x56G) ASIC for shipment in early 2021.  Cisco became the third vendor to announce a  25.6 Tbps ASIC this year, which we see as a higher volume class of ASIC than 12.8 Tbps ASICs.  At the same time, Cisco also announced a new line card on the Nexus 9500. Cisco's ASICs have ramped significantly in the Nexus 9K since the platforms market introduction and has continued to gain traction is the DC ASIC business as the Nexus 9K ramps in the Cloud as a strong alternative to Broadcom.
 
The company continued to push for 800G QSFP-DD, maintaining the companies position of DD as the form factor of choice for optics instead of OSFP as industry consensus on form factor is still fractured.
 
The Nexus 9K continues to strengthen the software features and offerings for Enterprise and Cloud customers, and we expect the Nexus 9K to expand more into the DCI space with 400 Gbps compared to previous iterations of the Nexus 9K.  We believe Cisco's tools around telemetry and security will make for compelling solutions in today's multi-cloud environment.