Longer-lasting servers will save Amazon almost $1 billion

Prolonging the life of servers in its data centres will save Amazon the thick end of $1 billion, according to the company’s latest earnings report.

The company claims the completion of a “useful life study” of its servers means it can continue using them for up to six years, up from the working life of five years that the company used previously. That decision alone means the company saves $900 million on its depreciation costs.

It’s the second time in recent years that the company has decided to extend the lifecycle of its servers. In 2022, it increased the life span from four to five years, with Amazon’s Chief Financial Officer Brian Olsavsky telling investors that the company had refined its software to run more efficiently on its hardware. “This then lowers stress on the hardware and extends the useful life, both for the assets that we use to support AWS’s external customers as well as those used to support our own internal Amazon businesses,” Olsavsky said in 2022, according to a report on The Register.

Server hardware obviously makes up a big part of Amazon’s AWS business, which continues to go from strength to strength, according to the latest financials. Revenue from AWS was up 13% year-on-year, with the cloud business now well on its way to achieving $100 billion in annual turnover. Its annual revenue of $90.8 billion means it could conceivably pass that milestone in the next financial year.

Related reading: What are servers?

More durable hardware

Amazon’s ability to squeeze more working life out of its servers reflects other research that suggests server technology is growing increasingly more reliable.

For example, in 2022, backup firm Backblaze produced a report showing that SSDs used as boot drives were proving to be more reliable than hard disks.

The report claimed that after five years of usage, SSDs had an annualised failure rate (AFR) of only 0.92%, whilst hard disks had an AFR of 3.55%. What’s more, hard disk failures rose sharply after four years in active service as a boot drive, while SSD failure rates remained flat even after five years.

All of this is not only good news for data centre operators, but good news for the environment too, with replacement cycles lengthening.

Additional Amazon / AWS coverage

• AWS’ strategic steps to close the GenAI competitive gap
• AWS pours $15 billion to expand cloud offering in Japan
• AWS Marketplace adds third-party services to manage your SaaS
• FTC’s case against Amazon is misunderstood
• Amazon takes gloves off in cloud computing probe