Sustainability and Tape Data at Scale
Every day the world is blasted with commercials and information talking about digital data. The growth of data, the protection of data, and the long-term durability of data are all hot topics. Digital data is critical to our daily lives. Yet for IT professionals, archive data, the 60-80% of data that is infrequently accessed, is almost an afterthought. The traditional methods of adding the same tier of storage are not sustainable within the tight IT budget.
But what if IT could fund digital transformation from Archive Storage?
According to IDC*, object storage and file systems are generating equivalent revenue, while the longevity of file systems represents ~80% of the unstructured storage market. Converged archive storage is fully integrated with existing unstructured data infrastructure. Converged archive systems enable tiering of unstructured data from active storage to archive tape, establishing true active archives. Let’s do the quick math, up to 80%2 of data qualifies as archive data (rarely accessed), 11- nines of durability3 tape is almost 16X less expensive than HDD storage. By “automagically” tiering rarely accessed data to tape, data storage administrators can save up to 90% of storage costs. That 90% is an investment in digital transformation, not storage savings!
Why are admins not implementing more tape?
In a word, education. Around 2010, the storage market began a modernization around object storage. As I previously pontificated on, the promise of object storage is that the efficiency would only ever require HDD. The developers were very smart regarding implementing software to manage the infrastructure with freely available code. But the math behind the hardware costs and durability of data requirements at an enterprise-level were grossly underestimated. From the point of view that object storage could handle billions of objects, the developers were right on, but without the ability to move the data to different tiers, costs have continued to rise. Tape is still less expensive and has higher data durability than HDD. This data durability is better regardless of the measurement used (BER or calculated Nines of durability).
Tape infrastructure is not taught in college and there are very few education programs covering usage, operation, and longevity. Humans tend to lean towards what they know or are taught in many instances. Even the most passionate active data administrator will agree that tape plays a key role in data storage management.
If you like the price, you’ll love the sustainability
Sustainability often comes at an increased cost. Not with active archives! Tiering data improves sustainability with every TB that is moved from HDD to tape. According to a study by Brad Johns Consulting, tape has an 87% lower CO2 footprint than HDDs. The linear density of tape is also continuing to grow without impacting the CO2 footprint. As of 2021 tape media is 20TB native and are targeted to be above 40TB within the next generation of tape, further driving the sustainability gap between tape and HDD.
CO2 calculations take into account energy consumption, data center scaling and consolidation are improving the CO2 impact, and adoption of tape in hyper-scale data centers is driving even more savings. Based on publicly available energy requirements published by IBM and others, the average 10PB tape library consumes 90% less energy than a 10PB HDD active archive. That difference becomes significant over the 5–9-year life of the data storage mediums.
A significant element of sustainability is the disposal and resulting e-waste. Tape media contains no hazardous materials; thus, destruction handling does not require special equipment, nor does it generate much e-waste specific to hazardous materials. The destruction of media for HDD requires specialized equipment (or a lot of physical labor with a hammer). Most companies destroy HDDs with specialized metal shredders requiring investment and management. Tape media can be easily shredded in the equivalent of a paper shredder, and the case of the media is 90% recyclable! Combine the effort of destruction and disposal with the average time of utilization for tape being 2-3 times longer and tape represents a significant benefit to our sustainability efforts.
My conscience is clear when recommending tape media for massive active archive data storage.
♦ Brad Johns Consulting LLC – Reducing Data Center Energy Consumption and Emissions with Modern Tape Storage
♦ IDC – Worldwide CO2 Emissions savings from Cloud Computing forecast, 2021-2024
♦ Marketsandmarkets – Next Generation Data storage market with Covid 19 impacts
♦ IBM tape TCO tool – https://www.ibm.com/it-infrastructure/resources/tools/storage-tco-calculator/