Data Centers – Part 3 – Pooled Infrastructure

One of the hot new growth areas in data centers is what I call “pooled infrastructure”.  The data center is a pool of interchangeable servers that can be rented out on a moment’s notice.  For example, a customer can rent out 30 servers during peak usage and 10 servers during off-peak usage.  Allowing customers to share a pool of servers greatly increases hardware utilization.

Along with the push towards renting out server capacity, the infrastructure providers are bundling value-added software with their infrastructure to save time administering and setting up the infrastructure.

Advertisements

Data centers – Part 2 – Centralized management

With traditional colocation, a data center will rent out space to many different customers.  Each customer brings in their own server hardware.  One of the trends among data centers is to have the data center purchase and manage all of the hardware.  This creates various economies of scale and benefits such as:

1. Standardization of hardware within a data center allows for elasticity and rapid provisioning of a server.  This opens up a new market for customers who want to rent servers.  Customers can have an operating serving within minutes or hours rather than weeks.
2. Software can automate the process of setting up servers.
3. Centralizing the purchasing and management of hardware reduces the inventory of spare parts that must be kept on site.  It also allows for repairs to have a much faster turnaround time because clients do not need to physically go to the data center to repair their servers.  These are very minor efficiencies.
4. If clients do not need to actually visit their servers, certain efficiencies are possible.  Vertical integration between data center design and server design allows for various capex and opex savings from power efficiency and cheaper servers.
5. Some infrastructure providers offer managed IT services.  They help customers to secure their servers, to maintain the server’s software, etc. etc.

Data centers – Part 1 – Overview

In data centers, the most interesting areas currently are in:

1. Software.  In particular, Amazon has been a pioneer in creating value-added software that helps software companies quickly setup databases and servers and backup storage.  Google, Microsoft, and IBM/SoftLayer are doing somewhat similar things.  The problem is that the data center segments of these companies likely will not move the needle much.
2. IT services.  Rackspace in particular is doing very interesting things.

The least interesting areas are companies that are mainly involved in selling space in a building because that business is a commodity.  Those companies will not  make a lot of money and they will not lose a lot of money.  While I think a lot of the data center REITs are overvalued and are suckering institutional investors into overpaying for their stock, they are not compelling shorts because they aren’t losing money quickly.  As well, many (legitimate) data center companies have been taken over at large premiums by companies like Verizon and Cogeco Cable at valuations I would disagree with.

*Disclosure: No positions.

Notes on cable – Part 2 – The future

I think that it’s likely the following trends will play out in the future:

• On-demand television will become widespread because it is the most compelling delivery mechanism.
• Technologies incapable of on-demand viewing (terrestrial broadcasting, satellite) will become less profitable.
• High-speed Internet providers (cable and telcos) will be able to exercise increasing pricing power over their customers.

The future of advertising

Summary:

• Online advertisers should be able to monetize their ads more effectively in the future.  This will be a huge tailwind.
• Scale will be a moat.
• Youtube has a bright future ahead of it.
• I suspect that television advertising and ad agencies will face serious headwinds.

What happens when technology is good enough?

There are some areas in technology that have reached maturity and are at the point of being good enough.  Take JPEG image compression for example.  This old technology is used on websites everywhere to compress images.  There is a newer JPEG2000 format which is technically superior.  However, this superior technology has seen very little adoption on websites.  Internet speeds are so fast that the benefits (webpages loading slightly faster) aren’t important enough for webmasters to switch.

For some shareholders, there may be a disappointing future ahead.  Once the demand for more computing power stops, there is no longer room to create value with newer technology.  Often what happens is that mainstream consumer demand goes away first.  This can devastate the business model of companies developing new technology as it can only be sold on a smaller scale to niche markets.

An investor’s guide to search engine optimization (SEO)

Search engine optimization is about getting a website to rank higher in search engines.  The goal is to increase web traffic without having to pay for ads.

This primer on search engine optimization is relevant to these stocks:

• RetailMeNot (SALE)
• Demand Media (DMD)
• Search engines (GOOG, Yahoo Japan, IACI/Ask.com, MSFT/Bing)
• Phone book companies (YLO.TO, DXM)
• Somewhat relevant to online e-commerce companies (AMZN)