My thesis
The purpose of this article is to evaluate Arista Networks (NYSE:ANET) positioning in the high-performance switch market, to look at its competitors, and to discuss how the firm can gain more share in the larger cluster. Workload for Artificial Intelligence (i.e. AI). We see Arista as well-positioned for double-digit growth each year over the medium term, leveraging operating leverage. My company’s evaluation will complement the writing and scenario analyzes will include further limitations of possible outcomes. While the business case is appealing, the stock appears to have a positive growth price. I rate the stock a hold: I await a better entry point.
Investment Review
The market for Ethernet network switches is estimated to be around $45 billion, with Cisco ( CSCO ) leading the way. 40% market share (if annualized from recent IDC data, see graph below). Arista is second with 10% share followed by Huawei with 8% share. Pushed away from US/EU critical infrastructure for security reasons in recent years. Other important firms include HP (HPE), juniper (JNPR), and the Chinese company H3C. The rest of the market is highly fragmented. Note that outside of the Ethernet solution, Nvidia (NVDA) offers a competitive offering with its InfinityBand switches, originally designed by Mellanox. While InfinityBand appears to have superior latency performance, the addition of AI large clusters may offer an advantage to new generations of Ethernet switches (200G to 800G): a sub-segment where Arista seems to dominate. Arista Management Estimates its addressable market. The number grew from $37bn in 2023 to over $60bn by 2027, resulting in a CAGR of 13%.
Arista was designed for high-performance networking and was built as Google’s first mainstream client in 2004. The company benefits from a successful team as founders Kenneth Duda and Andy Bechtolsheim are still active as board members. Jayshree Ullal joined Arista as CEO in 2008, with previous experience as a senior executive at Cisco in the data center branch. In her early career, she was a semiconductor engineer at notable US tech companies. All of them have a substantial stake in the company which makes them aligned with the interests of the shareholders.
Strong key competitive advantage lies in its operating system but also its ability to design superior architecture. The high programmability of its EOS allows integration of external silicon chips sourced primarily from Broadcom ( AVGO ) and partially from Marvell ( MRVL ). The ability to quickly leverage the industry’s best-in-class chips without the friction of integration gives Arista a competitive advantage over Cisco, which uses delayed integration of in-house chips (ASICs) and Broadcom’s chips. Is. Arista manufacturing is handled by outside contractors.
Today, it dominates the high-performance 100G/400G switching market with over 35% share ahead of Cisco (20%) and other smaller competitors such as Juniper, Nvidia (Mellanox) and Chinese companies H3C/Huawei/ZTE. . (outside the US). We can see in the following figure that the data center Ethernet switch market is expanding through high-speed segments (>100Gb per second), where Arista has an edge. The company has already shipped some 800Gbps systems that could be the next growth driver.
Meta ( META ) and Microsoft ( MSFT ) represented 46% of its $4.4bn FY2022 revenue, growing from 25% in 2021 and 32% in 2020. 70% of Arista’s revenue comes from cloud hyperscalers and data center providers. While this reliance on two customers may bother me, in the meantime, they are sticky and have growing CAPEX plans. Also, Arista’s ongoing development plants in an enterprise campus network can help diversify the margins of future revenue streams. The firm sees $750 million in campus sales by 2025, which should account for 10% of its revenue.
Arista’s recent progress looks promising. First, it partnered with cybersecurity firms such as Zscaler ( ZS ) and CrowdStrike ( CRWD ) to offer network-as-a-service solutions. NaaS represents a rapidly growing market as enterprise budgets shift away from firewalls and VPNs. In 2024, the firm will showcase a new 800Gps Ethernet backbone network capable of handling connections with 10-30k GPUs. The development of large language models for AI requires significant performance improvements. In fact, as stated in Nvidia GPU performance increases by 2X and cluster size by X4 each generation, resulting in an overall X8 speedup on each new Advanced AI model. The addition of large language models that require multiple parameters and leveraging large cluster sizes through thousands of GPUs can help Ethernet networking gain share over InfinityBand and NV-Link.
What value can we expect?
I largely aligned myself with management’s financial goals. I modeled a 16% revenue CAGR from 2024 to 2027, slightly above the 13% expected growth from the identifiable market as Arista should continue to gain market share.
On operating expenses: I see R&D/Sales ratio at 14% (EST 12%-14% seen by management), S&M at 6.5% (seen at 6%) and G&A less than 2% (inline ) went to The effective tax rate should increase gradually as deferred tax assets are used up. CAPEX should remain low as Arista is outsourcing its production. I expect a release of working capital in the coming years as the cash conversion cycle (record high) should moderate with further inventory deliveries (post-buildup). With this scenario, I find that the firm could generate: Between $3.4bn and $3.8bn between 2025-2026, leading to a cash flow yield of 5% (P/E 20X) to 5.5% respectively. (P/E of 18X) ) which seems reasonable.
Given this model, a terminal growth rate of +3%, and a WACC of 10%, I get a share price of $221/share, which is slightly below the current market price: lots of good news already priced in. There are mills. To give more perspective, I implemented a sensitivity analysis, varying the discount rate and mid-term growth.
Looking at relative value, we can see that Arista is priced at the same financial ratio as its peers, which is on the high end for data centers.
Balance Sheet Analysis
The company has a net cash balance sheet. Cash and short-term investment positions increased by $3bn to $4.5bn over the nine months. Stock repurchases have reached a total of $1.6bn since 2020 without any dividends. M&A activity during this period was modest: Arista appears to have all the building blocks it needs to grow organically. Looking at the financials, we see that Arista’s FCF margin on revenue reaches 40%: it’s a cash generator. So, I believe it has more room to expand its shareholder buyback program.
Technical analysis
We can see that Arista’s stock price is highly correlated with the peer group mentioned earlier. This is reasonable as its revenue exposure is significantly dependent on the data center segment. Arista’s YTD performance has been strong and driven by solid quarterly beats, not so much on speculation that AI will be a key driver.
Risks
The first risk that comes to my mind is related to the competitive landscape. While Arista has steadily gained market share over Cisco and Juniper, this trend may change if the firm does not always maintain the latest solutions. Also, hyperscalar cloud clients can further leverage their internal CAPEX by building “white boxes.” However I believe that the shift towards AI-driven workloads should strengthen the firm’s competitive advantage because of the need for best-in-class solutions (high stability and customization, low latency and error rates). A second risk relates to Arista’s customer concentration: Microsoft and Meta represent about 40% of its sales and the highest cyclicality of the firm’s results (which rely heavily on its data center costs). can persuade
Result
Two decades of investment in high-performance networking has positioned Arista Networks to take advantage of the growth of AI workloads. An experienced management team has built a powerful suite of connectivity solutions backed by a stable, adaptable operating system favored by data center operators. The company’s solid balance sheet and strong cash flow generation give me confidence in the company’s ability to continue to innovate and redistribute any excess liquidity to shareholders. What can we add? Well, all that comes with a price, and the stock’s current level forces me to stay on the sidelines. I rate the stock a hold: I await a better entry point.