jeudi 5 décembre 2013

Mellanox Badly Dented, But Not Done

The last 15 months have been brutal for Mellanox (MLNX), as this developer of high-performance switches, adapters, and circuits has been pummeled by a slowdown in the high performance computer space and a mass "abandon ship" call from growth and momentum investors. Mellanox certainly faces serious challenges, from the steep declines in computing to uncertain and unsteady growth in Web 2.0, cloud, and storage markets to competition with Intel (INTC). Even with that, though, I believe the underlying long-term market trends still favor Mellanox and I believe this beaten-down stock could trade back into the mid-$50s.


HPC Not DOA, But Certainly Not A-OK


It wasn't so long ago that Mellanox derived around 60% or so of its revenue from selling its Infiniband products and technology into the high performance computing market, particularly to OEMs like IBM (IBM) and Hewlett-Packard (HPQ). Unfortunately, the HPC market peaked back in 2012, with sales trending down 30% or more as Intel looks to the Grantley processor launch next year and major customers like HPQ work down


With that, Mellanox has moved from a go-go-growth story to a "where has the growth gone?" story; the shares trade today at one-third the price of the September 2012 peak and the average sell-side price target has plunged from over $100 to just under $50 today.


Consistent inconsistency is likely to be the order of the day for the foreseeable future in HPC. Intel basically controls the processor side and there is notable cyclicality between processor launch cycles. Likewise, there is a swirling process of evolution in the HPC space as enterprises reevaluate their real needs and the best way to meet them.


Storage, Cloud, Web 2.0 Coming, But Not Consistently


Mellanox does not rely just on the HPC market, even if it is far and away the largest component of its $5 billion to $6 billion addressable market today. Although companies like EMC (EMC) and NetApp (NTAP) have seen some turbulence in their respective parts of the storage world (and there are concerns that enterprises have overspent in storage and/or will need less by relying more on alternatives like Amazon (AMZN) Web Services), the melding of computing and storage, the adoption of flash storage, and the sheer growth in enterprise data mean a greater need for the high-bandwidth interconnections that Mellanox can provide.


Cloud and Web 2.0 are also emerging opportunities for Mellanox. Building on the relationship with EMC, Mellanox also works with VMware (VMW) and is the sole interconnect provider for that company's new cloud service platform. Microsoft (MSFT) has also been working closely with Mellanox, choose to go with Infiniband instead of Ethernet alternatives in various cloud and Web 2.0 offerings. All told, I think the Web 2.0 environment is starting to resemble the HPC market more in terms of its need for high-end interconnects, and that should be good news for Mellanox.


Competition Always A Threat, But Mellanox Already Looks Ahead


Maybe the biggest arguing point on Mellanox is the extent to which Intel is going to ruin their party in Infiniband and chew into the 80% or so market share that the company has long enjoyed. I'm skeptical that Mellanox is in serious danger. True, companies seldom ever hold on to that kind of market share in a lucrative market, but my opinion on Intel is that the company is much better at talking and projecting than delivering. Yes, Intel will likely grow its Infiniband business, but I think Mellanox is growing its addressable market and product/technology offerings fast enough to stay ahead of the problem (harkening back to the joke that you don't have to be faster than the tiger, you just have to be faster than your slowest friend).


Looking back over the last year or so, Mellanox has clearly been investing in the resources and technology it thinks it needs for the next big jump to 100GB/s. Buying IPtronics gave Mellanox high-speed semiconductor technology that will help it manage the gap between optical and electrical interfaces. Likewise, Kotura gives the company photonics technology that should position the company for the optical modules that will be important to 100G connectivity. There's also Voltaire - the acquisition of which gave Mellanox a more complete end-to-end array of solutions, as well as additional software assets.


It Doesn't Take Huge Growth To Make The Numbers Work Now


Given the ups and downs of the HPC market and the still-uncertain place for Infiniband in storage, Web 2.0, and cloud, Mellanox isn't likely to offer a smooth and consistent growth trajectory from here (not that any high-end tech companies can). I do believe, though, that there are positive underlying trends in the growth of demand for high-performance interconnect that will serve Mellanox in the years to come.


I'm looking for high single-digit revenue growth, as newer/expanding opportunities in storage, Web 2.0, and cloud offset some risk from Intel in Infiniband and broader competition in Ethernet. I'm looking for slightly less free cash flow growth (around 7%), as I do believe competition will eat away at some of the inherent profitability, but not to an overwhelming extent.


The Bottom Line


Putting those numbers into my DCF model, I come up with a fair value of around $53 to $54 for Mellanox - a pretty attractive premium to today's price. I'm certainly concerned about the risk that enterprises have over-invested in hardware and that key customers like IBM, HP, and EMC will not be able to reaccelerate their purchases in the next year or two. That risk, not to mention the competitive threat of Intel, goes a long way towards explaining why these shares may be cheap.


Mellanox is definitely not for everybody, and it's certainly not a buy-and-forget type of story. But insofar as I believe Mellanox has the best technology for functionality that is seeing increased demand, I think this is an interesting stock to consider today.


Source: Mellanox Badly Dented, But Not Done


Disclosure: I am long EMC. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. (More...)



This entry passed through the Full-Text RSS service — if this is your content and you're reading it on someone else's site, please read the FAQ at fivefilters.org/content-only/faq.php#publishers.






from SeekingAlpha.com: Home Page http://seekingalpha.com/article/1880191-mellanox-badly-dented-but-not-done?source=feed

Aucun commentaire:

Enregistrer un commentaire