Data warehouse appliance opponents like to argue that history is conclusively on their side. Database machine maker Britton-Lee, eventually bought by Teradata, fizzled. LISP machines were a spectacular failure. Rational Software’s origins as a special-purpose Ada machine maker had to be renounced before the company could succeed.
But the true story is more mixed. Teradata continues to this day as a major data warehouse technology player, and as far as I’m concerned Teradata indeed makes appliances. If we look further than the applications stack, we find that appliances actually occupy a large and growing share of the computing market. So a persuasive anti-appliance argument has to do more than just invoke the names of Britton-Lee and Symbolics.
I just ran across an article by MIT professor Samuel Madden that attempts to make such a case. And his MIT colleague Mike Stonebraker made similar arguments to me a few days ago. They are not wholly unbiased; indeed, both are involved in Vertica Systems. With that caveat, they have an interesting three-part argument:
- (Madden emphasis) Commodity hardware vendors have higher volumes, and hence can make more aggressive price/performance improvements over time than smaller specialists.
- (Stonebraker emphasis) What’s accelerated is only part of the task anyway. So the absolute ceiling on the benefit from hardware acceleration isn’t that great, even if part of the job were to become infinitely fast. Any price/performance gains created in this way are obliterated by the hardware margins appliance makers commonly charge.
- (Implicit, although neither went out of his way to call it out.) Small data warehouse appliance vendors may not happen to use the very best software strategies. And the advantages of the right software design can swamp any advantages to using specially accelerated hardware.
Frankly, I don’t find either #1 or #2 very persuasive. Vendors’ cost structures and vendor pricing are only loosely connected, given appliance vendors’ high gross margins. What’s more, those arguments miss one of the key points of appliances – hardware/software tuned in combination with each other.
So the pro/con argument for data warehouse appliances really comes down to #3: Are data warehouse appliance makers doing the best job of system architecture and implementation, or are software-only vendors doing better work? I don’t know of any way to judge that except by looking at particular products, vendors, and technical strategies – as we’ve been in the (highly overlapping) data warehouse appliance and data warehouse/ROLAP categories of this blog.