Wasabi Wipes Away the Tiers

I believe that “simple” always wins. That goes for simple-to-use appliances, simple cell phone pricing plans, and simple-to-use apps. The first generation of anything tends to be complicated and expensive. I recently upgraded my home entertainment system and, believe me, I was thrilled to finally say adios to that basket full of remote controls! My new Apple TV is so simple. I love it.

Cloud Storage Just Got a Lot Simpler

With respect to cloud storage, we’re just at the point where second-generation solutions like Wasabi are starting to appear. The first-generation products, mainly Amazon S3, Google, and Microsoft cloud storage, are all completely incompatible. You have to rewrite your applications if you want to move from one to another. And their pricing is so complicated that an entire cottage industry of consultants and third-party cloud cost monitoring companies has emerged to help you figure out what these services are going to cost you.

As I’ve written before, our vision is that cloud storage should be a one-size-fits-all commodity, like electricity or bandwidth. You shouldn’t need complicated pricing or tiers with different performance levels. After all, you don’t have multiple electrical outlets in your wall for different tiers of electricity. Cloud storage should be that simple.

What makes first-generation cloud storage so complicated?

1) Pricing tiers: $X for the first 1 TB, $Y for the next 9 TBs, $Z for the next 49 TB, etc.

2) Performance tiers: S3 Standard, Infrequent Access, Reduced Redundancy, Glacier, Nearline, Coldline, Normal Data Transfer, Fast Data Transfer, and on and  on.

3) Nickels and Dimes: You pay for the storage, but then there are a host of other little charges for egress, API calls (PUT, GET, LIST, etc.), data retrievals, lifecycle transition requests, and ad infinitum.

It gets worse. Here’s a screenshot of Amazon’s so-called “Simple Monthly Calculator.” Guys, just because you call it “simple” doesn’t mean that it is actually simple. In fact, the reason that you need a calculator in the first place is because the underlying price list is so complicated. You have tiered pricing by storage type, tiered pricing based on volume, pricing based on how many times you touch the data, tiered pricing based on the amount of data you take out, tiered pricing based on how fast you want to get your data, and the list goes on.

The pricing page for Amazon cloud storage goes on for 9 pages!

A simpler way to calculate cloud storage costs

Here’s my idea of a simple calculator: it’s called the multiplication symbol, and you only need one of them! Here’s how you compute what it’s going to cost you to store 500 TBs of data for a month in Wasabi:

500 x $6.99 = $3,495

• No usage tiers. There's only one price per GB. Everyone gets the lowest price.
• No performance tiers. You don’t need them because Wasabi is faster than Amazon’s fastest storage but priced like cold storage.
• No additional fees. There are no charges for egress, PUTs, GETs, or any of the other API request or transaction fees on Amazon’s list.
• The price is the price. Your Wasabi bill has exactly one line-item: Storage.

Performance tiers carry additional hidden costs beyond what shows up on the price sheet. Amazon and Google both have expensive “hot” storage and cheaper “cold” storage products. Consultants make a big deal about how much money they can save you be optimizing when you should move data out of expensive fast storage and into cheaper slow storage. In fact, at a recent Gartner conference, a storage analyst said that any enterprise thinking about migrating their data to the cloud should count on having one full-time employee dedicated to cost-optimization strategies. Think about that for a moment: one full-time employee just to figure out when to shuffle data around from one performance tier to another!

Why are pricing and performance tiers such a problem for CIOs?

Organizations have budgets. They need to know how much things are going to cost before they get the bill. If you’re coming from a world where you stored all your data in-house, you never had to worry about things like egress, API operations, retrievals, and the like. In fact, you probably had no way to measure these things even if you wanted to. Now you’re thinking about moving your storage to the cloud and are trying to predict how much it’s going to cost you. But you have no way to accurately predict your usage thanks to all these different tiers and nickel-and-dime charges. It’s a real problem and people hate it.

Extra fees add up

These little charges can add up big time. Referring to Amazon’s price list, the basic cost to store 500 TBs in S3 is $11,050 per month (the first 50 TBs @$23 per month, and the remaining 450 TBs @$22). That comes to $22.10 per TB per month.

But that’s only if you never use even one byte of your stored data!

Access 1% of your data per month and the cost goes up to $23.50 per TB. Access 10% and your costs rocket to $30.01 per TB per month. That’s $15,000 a month for 500 TB of data–a 36% increase over the basic storage costs!

The comparable cost for storing 500 TB in Wasabi is $2,995 per month, irrespective of usage.

Why pay for your own data?

The unpredictability of costs is a big deterrent to cloud storage adoption, which is the main reason why we decided to get rid of egress fees. Plus, it just doesn’t feel right to be told that I have to pay extra to use my own data.

Instead of seeking out creative new ways to use my data, I’d be thinking “How can I avoid using my data and racking up unbudgeted expenses.”

It will take a few more years, but I predict that most pricing and performance tiers will eventually go away. You shouldn’t have to choose between fast and slow, high redundancy or low redundancy, and standards-based or proprietary.

And you certainly shouldn’t have to pay to get your own data back.