Cost-Optimized Data Protection and Disaster Recovery Strategies
Today’s rapid, unpredictable changes in server and storage system prices require organizations rethink how they purchase these products. Those organizations with flexible budgets may find they can pay up and get the servers or storage systems they need. However, many more may need to explore alternatives to meet their disaster recovery (DR) objectives to navigate today’s SSD shortages.
AI Driving Hardware Shortages and Higher Prices
Today’s CPU, memory, and storage component shortages primarily stem from hyperscalers building new data centers to host artificial intelligence (AI). One research firm put hyperscalers’ capital expenditures (CAPEX) for AI data centers at $630 billion in 2026. They also estimated the total global AI data center CAPEX for all companies will grow to $8.9 trillion from 2026-2030.
Filling up these data centers with the necessary infrastructure has specifically created shortages of DRAM and SSDs. The price of DRAM alone has increased by up to 90 percent quarter-over-quarter while SSDs have experienced similar cost increases. Adding insult to injury, organizations cannot assume they can obtain servers or storage systems even if they pay higher prices.
More hardware technology providers now issue quotes that expire in 30 days or less. One provider that spoke with DCIG shared it now resets its storage system prices at the start of each month. Another provider said it will no longer publish the type of CPUs its storage systems use. Further, servers and storage systems with the latest, most powerful or highest capacity technologies will command substantial premiums.
These shortages coupled with ongoing price increases may require organizations to explore alternative technologies to meet their DR objectives. In some cases, media such as disk, tape, or even refurbished SSDs may again find new roles in their environment. In other cases, organizations may want to utilize software they already own to meet these needs.
Disk and Tape not Dead Yet
Many organizations started using only SSDs for primary storage and only hard disk drives (HDDs) for secondary storage for both performance and cost reasons. SSDs performed significantly better than HDDs while HDDs offered similar performance benefits over tape. While SSDs cost more than HDDs and HDDs cost more than tape, their performance benefits outweighed their respective costs.
While the performance delta between SSDs, HDDs, and tape remains the same, the cost delta has widened significantly. Technology providers now report SSDs once again cost as much as 20 times more than HDDs. The price gap between HDDs and tape has also similarly widened.
As a result, organizations may find they may need to expand their use of disk and perhaps even tape. This adoption does not mean they necessarily need to abandon using SSDs in production or HDDs in secondary storage.
Rather, they can deploy hybrid storage solutions which use both SSDs and HDDs in their production servers and storage systems. These solutions keep the most active data on SSDs and tier older or less active data to HDDs.
Similarly, for external secondary storage, organizations can possibly use HDD-based systems as backup targets and for fast backups and restores. These HDD storage systems may then offer the option to tier older backups to cloud storage or tape. This configuration permits organizations to meet their more aggressive backup and recovery requirements. They can then tier backups to lower cost media to meet their compliance or retention requirements.
Life Left in Old SSDs
Those organizations that still want or need SSDs but cannot afford them may still have a path forward. One technology provider recently shared, many SSDs on the secondary market still have lots of life left in them.
Unlike HDDs, SSDs have no moving parts. So, one cannot necessarily measure the life of an SSD by its age. Rather, one determines the lifespan of an SSD by its usage or wear life. SSDs only experience “wear” when applications write to them, not when applications read from them. As a result, SSDs which were infrequently written to when initially deployed may still have significant life left in them.
Before using SSDs available on the secondary market, organizations should verify at least two technical steps have occurred.
- First, verify the reseller of the secondary market SSDs wiped the SSDs to remove all data.
- Second, verify how much wear life remains on each refurbished SSD as remaining wear life will vary. Software exists that resellers can use to determine and document the remaining wear life of each SSD.
Using SSDs available on the secondary market also presupposes an organization corporately permits the purchase of them. Some organizations do not allow purchases from the secondary market due to regulatory or support concerns. However, organizations that can buy from the secondary market may find ample SSDs available at a substantially lower cost.
Optimizing Available Hardware Resources
However, many organizations remain committed to using SSDs and buying SSDs on the secondary market may not be an option. At the same time, the current SSD shortage and their higher prices may force organizations to explore alternative solutions. Here again, multiple options exist for them to optimize their existing available hardware resources. In the remaining cases, this involves using features in software organizations may already own.
Server Virtualization-based Replication
Organizations already running server virtualization platforms can reduce disaster recovery costs by leveraging their built-in replication capabilities. These platforms often offer native replication features that asynchronously copy virtual machines (VMs) to a secondary site.
This server virtualization software may already include the replication software license or make it available for an additional cost. More importantly, the software can ideally utilize an organization’s existing storage resources to minimize or eliminate the need for more storage.
This software specifically includes policy-based automation that makes it easier to implement and use. Using policy-based automation, organizations may define replication intervals, failover priorities, and recovery workflows without deploying additional third-party tools.
Using these native replication capabilities, organizations can offer DR for their Tier 2–3 applications with moderate recovery point and recovery time objective requirements. Admittedly, the replication functionality on server virtualization software may not match the features on purpose-built replication software. However, organizations can balance protection, simplicity, and cost efficiency without making significant capital investments to extend their DR coverage.
Backup Software with DR Orchestration
While server virtualization software includes replication features, some organizations hesitate to use them. This hesitancy stems from many organizations moving to multi-hypervisor environments. As a result, they must learn how to implement and use the replications features across multiple server virtualization software offerings.
To overcome this concern, organizations can use modern backup platforms which include DR orchestration features they may already own. For instance, these platforms enable instant VM recovery. Using this functionality, organizations may run workloads directly from backup without requiring duplicate production storage.
They can also automate the creation of failover runbooks and perform non-disruptive DR testing using sandbox environments. These capabilities reduce the need to maintain fully provisioned secondary infrastructure while ensuring recovery processes remain validated and reliable. These features further minimize the need for organizations to maintain excessive storage capacity dedicated solely to disaster recovery scenarios.
By extending their use of these existing tools, it eliminates the need for them to acquire separate DR platforms. It also reduces their operational overhead by automating tasks which often require routine administrative oversight and intervention. Organizations seeking unified backup and DR capabilities without procuring additional storage resources should find this approach a good fit.
Beating the SSD Crunch: Cost-Effective Strategies for Resilient Data Protection and Disaster Recovery
Rising SSD prices, now up to 20-times higher than HDDs, require organizations to rethink their data protection and DR strategies. Rather than overinvesting in premium storage, enterprises should adopt hybrid models that tier active data on SSDs to HDDs. They should also explore shifting their storage of backups and less critical data to cloud, HDDs, or tape.
Organizations can also extend the life of existing assets and reduce new purchases. Secondary-market SSDs may still offer usable lifespan if properly validated for wear and securely wiped. Additionally, leveraging built-in replication and backup orchestration tools can reduce reliance on duplicate storage systems. In so doing, organizations can maximize their existing infrastructure before approving new capital expenditures.
To respond effectively, organizations should take three immediate steps:
- Implement hybrid storage tiering across production and backup environments.
- Evaluate refurbished SSDs where policy allows, and
- Activate or utilize existing native replication and backup orchestration features.
These actions will help organizations balance cost, performance, and resilience while maintaining effective disaster recovery coverage.
