S1:Backup

Backing up to 18TB Hard Disk Drives

May 26, 2022 by StorONE

IT professionals looking to reduce backup infrastructure costs are looking into backing up to 18TB hard disk drives. Since introducing 20TB hard disk drives, 18TB drives now own the price per TB advantage versus other hard drive densities, making them a compelling choice to use in backup storage targets.

IT has to make sure, however, that its efforts to save money don’t hinder the number one responsibility of the backup process, fast recoveries. Fast recoveries are more than just the performance of the underlying backup storage media. The solution needs to support modern backup server software’s rapid data transfer capabilities. Backup storage targets need to maintain a high ingest rate even if they experience a drive or controller failure.

The Problems Backing up to 18TB Hard Disk Drives

Today’s major hard disk manufacturers deliver hard drives of 16TB, 18TB, and 20TB, with 32TB hard disk drives on the horizon. However, many legacy backup storage vendors only allow their customers to use 8TB drives, a density introduced almost seven years ago, even though 18TB drives are only about $100 more per drive.

There are two primary reasons these legacy vendors force customers to use such old technology and pay so much extra per TB. First, these systems need many spinning drives to deliver any reasonable level of performance. Using 18TB drives means cutting the required hard disk drives in half, dramatically impacting the performance of legacy systems. Second, legacy backup targets using 8TB drives require three days to recover from a failed drive (RAID rebuild). During the multi-day drive recovery process, IT must choose between slowing backups down so the rebuild can happen faster or slowing the drive rebuild process, which maintains good backup performance but risks losing the entire backup dataset to an additional drive failure.

Is All-Flash Backup an Alternative?

There is a new breed of backup storage vendors promoting all-flash backup storage. While all-flash backup storage does improve backup ingest rates and is less likely to experience a multi-day drive rebuild, an all-flash system is flash-only. Despite what all-flash vendors say, the raw cost of flash media is about 10X the cost of hard disk media. That’s $3,500 per 16TB flash vs $350 for 16TB of HDD.

Deduplication isn’t a key differentiator for all-flash backup. If you are going to use it, deduplication is universally available on both hard disk drives and flash drives. Fast recovery from drive failure won’t be a key differentiator for long either. As flash drive densities increase, so will the time it takes to execute a RAID rebuild. We are already hearing reports of eight to twelve-hour recovery times.

Backing up to 18TB hard disk drives without compromise

The economics of 18TB or larger drives are compelling only if IT can adopt them without compromising the recovery effort. Achieving this balance requires three advancements in technology that are unique to StorONE. First, we rewrote and optimized the storage IO stack, which means we can move data through our infrastructure as fast as possible. Second, we rewrote erasure coding from the ground up so that volumes built from 18TB hard disk drives can recover from a drive failure within three hours instead of three days. Third, we rewrote auto-tiering to deliver consistent performance, even with high-ingest workloads like backup storage.

The rewrite of auto-tiering enables IT to use a low-drive-count but high-performance flash tier and the granular backup capabilities of modern backup software to receive dozens if not hundreds of simultaneous backup jobs continuously throughout the day. This frequent capture of production data means IT can deliver the best recovery point objective (RPO) possible. The rewrite and optimization of the IO stack allow StorONE’s S1:Backup to deliver 300K IOPS+ from just a handful of flash drives, keeping costs low.

The auto-tiering capabilities then predictively and proactively move older backup jobs to a large hard disk tier, made up of 18TB or greater hard disk drives. S1:Backup does not wait for the flash tier to fill up. It is constantly looking for opportunities to move older data from the flash tier so that there is always room for the next inbound backup job. The result is cost savings and consistent performance.

Backing up to 18TB hard disk drives during a rebuild

The final challenge is recovering from a failure of one of these 18TB drives. StorONE’s rewrite of erasure coding, combined with its rewrite of the IO stack, enables us to deliver sub-three-hour drive recoveries. Recovering from a failed drive requires three key steps. First, the process needs to read all the drives to collect the parity bits associated with the failed drive. Second, the process needs to run a complex mathematical parity calculation to determine what data is now missing due to the drive failure. And third, it needs to write the data on the failed drive to the remaining drives. Our development investment enables us to complete that process faster than any other vendor on the market, as seen in this video.

While the record-setting three-hour rebuild is occurring, backups continue at full performance because they land on the flash tier first, so there is no interruption to the backup process. If there is a flash drive failure, those drive recoveries happen within minutes.

Fast Recoveries with 18TB hard disk drives

S1:Backup efficiently uses 18TB hard disk drives within the backup storage target thanks to StorONE’s three unique development efforts, rewriting the IO stack, erasure coding, and tiering. The small flash-tier speed ingests enable the best possible RPO. These three outstanding development efforts also help deliver the best possible recovery time objective (RTO).

Most modern backup solutions now have a recovery-in-place (instant recovery) feature that instantiates a virtual machine’s or an application’s data directly on the backup appliance. There is usually a significant performance delta between the backup storage target and production storage. With StorONE, instantiation occurs automatically on the flash tier, delivering a production-class performance experience.

Thinking Results

The result for our customers is the best possible RPO/RTO while reducing backup storage costs by 50% or more. The use of these high-density drives also means that the S1:Backup solution consumes about 1/3 the rack space of competing solutions, consumes far fewer network ports, and reduces power consumption.

There will come a time when 20TB drives have the price-performance advantage over 18TB HDDs, probably when 22TB or 26TB drives come to market. With StorONE’s technology, you can mix in the next generation of HDDs and use them to their total capacity on the existing volumes, eliminating migrations and storage refreshes.

To learn more about reducing backup storage costs, watch our live webinar “TechTalk – Four Myths that Increase Backup Storage Costs”“

Optimizing Veeam Backup Storage

May 19, 2022 by StorONE

With innovations like change block tracked backup, a scalable repository, and instant recovery, IT professionals should prioritize optimizing Veeam backup storage. Without proper optimization of the backup architecture, IT can’t realize the full potential of Veeam’s innovations.

Veeam Backup Storage Needs Flash

The granular data movement capabilities of Veeam’s change block tracking (CBT) backups enable customers to increase backup frequency which should improve recovery point objectives (RPO). The problem is that while CBT backups are easy on the network, they require a lot of random IO performance from the backup target.

The IO profile of CBT IO profile is highly randomized and needs a storage system to respond to that randomization. Hard disk technology can’t keep pace with dozens or even hundreds of simultaneous backup jobs landing on the backup storage target every hour. Flash storage, however, can handle these high-frequency backups, and it should be a component of every backup storage target.

Veeam Backup Storage Needs Hard Disks

While flash storage is critical to improving RPO, we are not at a point where an all-flash backup storage target makes sense. Flash is still too expensive to compete with hard disk drives. An 18TB hard disk drive is 1/10th the price of a 15TB flash drive.

Most legacy vendors shy away from using high-density hard disk drives for fear of customers’ slow recovery from drive failure complaints. We repeatedly speak to IT professionals that tell us they are experiencing RAID rebuild times over three days with 8TB drives, let alone 18TB drives, which will take even longer.

Veeam Backup Storage Needs Flexible Immutability

Veeam is doing an excellent job advancing state-of-the-art ransomware detection and protection. However, you can never have enough immutability. Protection from ransomware attacks requires instant immutability across all protocols without transferring to another storage silo. You also need to make sure any instantly recovered virtual machines or applications can be protected until you can ensure the environment is clear of any remnants of the malware.

You Need Production Class Recovery

Veeam’s instant recovery feature is a critical innovation that enables a virtual machine’s data to be instantiated on the backup storage target, eliminating network transfer time. The problem is that legacy backup storage devices don’t have the performance or availability to provide a production-class environment, forcing IT to move that VM into production quickly. Also, if the data center is experiencing a ransomware attack, instantly recovered VMs data is exposed to reinfection.

StorONE’s Backup is Optimized for Veeam

StorONE’s S1:Backup is optimized to elevate Veeam’s innovation. Its flash-first technology enables the rapid ingest of data. You can improve RPO by executing dozens of simultaneous backup jobs. However, StorONE’s S1:Backup is not flash-only. We intelligently integrate flash and hard disk drives to deliver performance and affordable long-term cost savings.

Optimizing Veeam Backup Storage — S1:Backup Optimized Veeam Backups

S1:Backup provides immutability within seconds of the completion of each backup job. There is no data transfer to an immutable secondary volume or a secondary storage system. While S1:Backup supports S3, you can also enjoy immutability across all block and file protocols.

Finally, S1:Backup provides a production class recovery environment within the same storage system. During a recovery, flash is dynamically reallocated so that capacity is available to host instantly recovered VMs and applications. The solution has built-in high availability and provides the same immutable protection of the instantly recovered VMs that it does for backup data.

StorONE, ONE backup solution for Veeam

StorONE rewrote the IO stack, and we own the entire IO path as data travels through our platform.

Optimizing Veeam Backup Storage with S1:Backup also provides 360° Ransomware Protection — S1:Backup Provides 360° Ransomware Protection

We also rewrote erasure coding to deliver the fastest recovery from drive failure. We even optimized tiering to accommodate the backup use case.

Combining these innovations means StorONE can deliver consistently high performance from a handful of flash drives, affordably meeting your RPO and RTO demands. It also means you can safely use high-density hard disk drives as our vRAID recovers from a drive failure of even the highest density hard disk drives in less than three hours. You can also seamlessly integrate new higher density drives without needing to create new volumes and reconfigure backup jobs. Optimizing Veeam Backup Storage with S1:Backup also provides 360° Ransomware Protection.

See You At VeeamON 2023!

Join StorONE at VeeamON May 22-24 in Miami. We’ll be at booth S2 and are pre-booking demos. Click here to reserve your 1 on 1 demo time.

You can also watch our webinar to learn more: “Four Ways to Optimize Veeam Backup and Recovery.”

The ROI of Fast RAID Rebuilds

April 28, 2022 by StorONE

Limiting the impact on production performance, improving capacity utilization, and enabling high-density hard disk drives are just a few examples of the ROI of Fast RAID Rebuilds. RAID rebuild time is the time it takes for a storage system to return itself to a protected state after a drive failure. The time window is critical because if another drive fails during this time, there may be complete data loss. Also, the rebuilding effort on most storage systems will significantly impact production performance. Slow rebuilds require IT to take extra measures, which cost money, to limit the organization’s exposure.

The Anatomy of Fast RAID Rebuilds

To improve RAID rebuilds, most vendors focus on the media, forcing customers to use either flash drives or capping the density of the hard drive their systems use. The problem is not the media and, with the correct algorithm, it has a minor role to play in the rebuilding effort.

Fast RAID rebuilds require three fundamental elements. First, the storage solution needs to be able to read data from the surviving media as fast as the media can deliver it. Second, the storage solution needs to execute a complex mathematical operation to determine what data was on the failed drive via parity. In straightforward terms, it is solving for “x.” Then finally, the storage solution needs to write data as fast as possible to the surviving drives.

Most legacy storage solutions, be they all-flash, hybrid, or hard disk only, don’t own the IO path. They leverage open source or public domain file systems and code to get to market quickly. They could not improve IO performance to a drive even if they had the capability. As a result, most “improvements” in RAID rebuild times occur because the vendor is using faster media, so all-flash systems are mistakenly seen as a fix for long RAID rebuild times. The vendor didn’t change their legacy code (because they can’t); they just leveraged a faster technology.

Designing Fast RAID Rebuilds Algorithms

The first step in fast rebuilds is to make sure a drive failure protection algorithm is in place which enables the system to read and write data from all the surviving drives. Most modern RAID algorithms do read data from all available drives, but the overwhelming majority do not write data to all surviving drives. In most cases, these systems have to set aside (reserve) drives, called hot spares. During a rebuilding process, they will activate one of these hot spares, and all the recalculated data is written to that single drive, which will bottleneck the recovery process.

The data protection algorithm also needs to be intelligent in determining what data is missing from the drive. Most algorithms are only granular to the drive level, not the data level, which means a 20TB hard disk drive with 1TB of data on it is as slow to recover as a 20TB drive with 19TBs of data on it.

The algorithm not only needs to intelligently use surviving drives and intelligently determine which data needs to be recreated, but it also needs to be efficient so its execution doesn’t impact production performance while the rebuilding process is executing. Ideally, the entire storage software component is efficient, so there is plenty of available CPU power to execute the rebuild algorithm while still having plenty of resources to maintain application performance.

Read and Write Performance Impact Fast RAID Rebuilds

Armed with an algorithm that enables all surviving drives to be read and, more importantly, to be written to during the rebuilding process also enables fast RAID rebuilds. Again, the problem is most legacy vendors don’t own the IO path to the media and can’t do much to improve the read and write performance of their storage systems other than forcing their customers to buy faster and more expensive media. If they continue to innovate, vendors that own the IO path can significantly improve IO performance and enable the software to extract a much higher percentage of the media’s theoretical maximum.

IT must look for vendors that own the IO path and the drive failure protection algorithm. Ownership of both the IO path and the drive failure algorithm means the vendor can coordinate the effort to deliver the fastest possible rebuilds in the shortest possible amount of time. The result should be hard disk drive recovery times of a few hours and flash recovery times of a few minutes while delivering high-performance to the organization’s applications and users.

Fast RAID Rebuilds Save Money

A storage software solution that addresses all three elements of drive recovery can enable customers to lower costs and accelerate the return on investment of the storage system. A sub-three-hour recovery time from a hard disk failure means customers can use the newest high-density hard disk drives without fear. A 20TB hard disk drive, is more than 10X less expensive than its 16TB flash equivalent.

Slow recovery times also lead most customers to increase their drive redundancy settings. The concern is if a second drive fails while the rebuild process is executing it can result in data loss and significant downtime. A multi-day recovery effort increases those chances of data loss. Dual parity RAID is commonplace today, and triple parity RAID (three simultaneous drive failures) is gaining in popularity. The problem with increasing drive redundancy is it lowers utilization levels and, in most cases, with legacy RAID algorithms, negatively impacts performance.

A modern storage solution should provide the flexibility to select as many drives for redundancy as IT requires without impacting performance. However, fast recovery times enable organizations to allocate fewer drives to redundancy. There is significantly less chance of multiple drives failing during a three-hour recovery than multiple-day recovery. These solutions should also execute the rebuilding process without impacting production performance. A modern algorithm enables IT to tailor drive failure protection on an application basis to balance protection and utilization properly.

The modern storage algorithm manages data, not drives, and is abstracted from the underlying hardware. It provides the additional benefit of flexible expansion, which eliminates storage refreshes and data migration. As both flash and hard disk densities continue to increase, the modern storage algorithm can integrate new, higher density drives and use them to their full capacity without requiring new volumes, forcing IT to migrate data from old to new.

Learn More

StorONE spent its first eight years rewriting the foundation of storage IO and building its storage engine, optimizing IO for maximum efficiency. With StorONE, you can extract more performance and achieve greater scale in a much smaller, more cost-effective footprint.

Click to learn more about our Enterprise Storage Platform or our advanced drive failure protection technology, vRAID.
An area where our optimized engine shows immediate value is backup storage; join us live next week for our webinar “Busting the Four Myths that Increase Backup Storage Costs.”

Simplifying Backup Storage Scaling

February 9, 2022 by StorONE

Given its never-ending growth rate, simplifying backup storage scaling is a critical requirement for IT to address. Today, backup storage is typically 5x the size of production storage. The need to recover faster, retain data longer, and repurpose backup data will increase to 10X production storage capacity. IT has to find a better way to address this growth to keep costs down.

The Complexity of Backup Storage Capacity Scaling

Legacy vendors will point to scale-out backup storage as the solution for scaling backup storage. These architectures add capacity by adding nodes to an existing cluster of storage. When the backup infrastructure needs more capacity, “just add a node.” While adding a node to the cluster may sound simple, this approach has severe logistical challenges.

Most nodes require 2U or 4U of rack space. Finding room for these additional nodes becomes a problem as the backup environment scales. When IT racks the new node, they must also supply network connectivity, typically four connections. The node also needs power and will require additional cooling. Then depending on the cluster design, IT may still have to reroute backups to that node.

Scale-out backup storage, in most cases, requires fixed capacity and a fixed device type per node. A typical configuration is limited to 8TB hard disk drives within each node. The 5X backup data multiple means that even small customers often need 1PB or more of backup. We find StorONE’s S1:Backup customers commonly asking for 3PBs or more. Scaling to 3PB may take 32 or more nodes in the typical scale-out configuration. As a result, most vendors limit the total capacity of the cluster to less than 5PBs.

An alternative is to use scale-up solutions, but these backup appliances use the same inefficient legacy storage software as most scale-out solutions. They typically only scale to 500TB before suffering a significant drop in performance. The 3PB requirement requires six independently managed storage systems.

Simplifying Backup Storage Capacity Scaling

StorONE builds S1:Backup on our enterprise storage platform, which uses the efficient StorONE IO Engine to deliver massive scale and performance from a minimal amount of hardware. StorONE can scale to 15PBs of capacity, without performance impact, in a single two-node cluster. With S1:Backup, IT can continue scaling the initial nodes to 15PBs. Instead of inflexible upgrades, IT can add higher density drives as they become available, mixing them into the same system without creating new volumes. StorONE’s S1:Backup fully supports 20TB drives, including RAID rebuilds in less than three hours.

The Complexity of Backup Storage Performance Scaling

The scaling of the backup data set results from the growth in production, application, and user data. More data puts more pressure on the backup storage targets to ingest data. The sensitivity to data loss and the ever-increasing ransomware threat motivates customers to backup multiple times per day to lower recovery point objectives (RPO). Scale-out backup storage vendors claim to keep RPO consistent as you add production capacity. As you add nodes to a scale-out backup solution, you, in most cases, will select a group of backup jobs and redirect them to the new node. Adding nodes, again, increases complexity, as does retargeting backup jobs. The scale-out architecture forces you to pay for more and more hardware because it runs on a legacy storage engine and cannot drive each node to its maximum efficiency potential.

Simplifying Backup Storage Performance Scaling

StorONE’s S1:Backup delivers maximum performance from the existing networking and storage interfaces and has plenty of bandwidth to handle more data transfers. Our Flash-First architecture ensures that the physical storage will keep pace with network transfers. The StorONE IO Engine is so efficient that it has excess CPU, Memory, and storage connectivity bandwidth available to receive more data. Still, if more ingest performance is required, the StorONE design means IT only has to add one network card instead of the cost of an entire node (server, CPU, memory, power supply, etc..).

The Complexity of Use Case Scaling

Most backup storage systems solve only one problem, storing backup data. The challenge is there are other storage needs both within the backup infrastructure and beyond it. Within the backup infrastructure, IT needs a solution to ingest the backups rapidly, a solution to store the backup data long term, a solution to securely store backup metadata, and there is an emerging need for a sterile, production-class recovery area.

Beyond backup, use cases like Archive and Network Attached Storage (NAS) are obvious next steps for a more flexible solution. Again, most backup storage solutions don’t address the backup use cases’ requirements. And they certainly can’t address more production use cases like VMware, Hyper-V, KVM, and bare-metal database servers.

Simplifying Use Case Scaling

No backup solution other than S1:Backup meets all the backup use case needs. It eliminates the need to create multiple storage silos within the backup infrastructure, which compounds the storage system sprawl throughout the data center. With S1:Backup, you gain confidence in the solution. As the legacy storage systems supporting other use cases “age-out,” you can integrate those use cases into the existing StorONE solution, leading you on a path to complete storage consolidation.

Conclusion

The traditional solutions to backup scaling are scaling out by adding nodes or other independent scale-up storage systems. The fundamental mistake in both approaches is not the technique; it is the underlying core storage architecture. These systems count on legacy storage IO stacks that don’t deliver the full potential of the individual hardware components. Cost-effectively scaling backup infrastructure requires a new storage IO engine that extracts the full potential from each hardware resource. The result is a more affordable solution, which is easier to manage, support, and upgrade.

Learn More:

Join StorONE on February 24th at 11:30 am ET / 8:30 am PT for our live webinar “Fixing the Three Inefficiencies of Scale-Out Backup Storage.” To learn how to overcome scale-out backup storage’s three inefficiencies:

Inefficient Use of Data Center Floor Space
Inefficient Use of Hard Disk and Flash Innovations
Inefficient Upgrades and Future Readiness

A Complete Ransomware Recovery Strategy

February 3, 2022 by StorONE

One of the challenges IT faces when developing a complete ransomware recovery strategy is to think holistically about the process. Focusing on just one capability may expose the organization to ransomware damage in other areas. IT needs a 360° Ransomware Recovery Strategy that protects production data backup metadata and rapidly returns the organization to an operational state.

The Elements of a Complete Ransomware Recovery Strategy

Frequent Backups
Immutable Storage
Rapid Recovery
A Sterile Recovery Environment
Affordability

Ransomware Recovery Requirement 1: Frequent Backups

For years (decades?), backup vendors have claimed: “it is all about recovery!” While recovery is a critical part of a ransomware recovery strategy, it is not all about recovery. If your architecture doesn’t enable you to execute fast, frequent backups, your recovery point objective (RPO) may lead to losing hours and even days’ worth of data. This delay may lead to a temptation to pay the ransom instead of completing the recovery process. A complete ransomware recovery strategy requires the frequent protection of production data. Also if that data isn’t written securely to persistent media, there may not be quality data for recovery.

Ransomware Recovery Requirement 2: Immutable Storage

Immutable storage is critical to your ransomware recovery strategy. Your backup storage target should store every backup in an immutable state and be able to retain that state for months without impacting performance. It should also store your backup software’s metadata in an immutable state, which means your backup storage target needs to deliver the performance your backup-server software requires to update these files quickly. If the ransomware attack encrypts either the backup data or the backup metadata, you might not be able to recover anything. A complete ransomware recovery strategy requires immutable storage of the protected data and the backup configuration files and indexes.

Ransomware Recovery Requirement 3: Rapid Recovery

Frequent backups and protected storage of that data set the stage for a successful recovery. However, these recoveries have to be some of the fastest IT has ever done since they are under pressure from users and application owners who want to get back to a productive state as quickly as possible. There is also pressure from the executive team, weighing the cost of paying the ransom versus the time it will take for IT to complete the recovery. Finally, there is the possibility that the “bad actors” have set a time limit for you to pay the ransom, or they will delete all your data. A complete ransomware recovery strategy requires that instant recoveries or even native recoveries occur quickly. Backup storage needs to provide the ability to host production data with a performance profile similar to that of the production environment.

Ransomware Recovery Requirement 4: A Sterile Recovery Environment

Unlike a natural disaster, the state of your production storage after a ransomware attack is at best “unknown.” Ransomware trigger files often make copies of themselves as they encrypt your environment. If IT doesn’t have the time to complete the forensic work, which can take hours, to ensure the removal of these files, then the likelihood of reinfection, with even less “pay the ransom” time, is very high.

A complete ransomware recovery strategy requires a sterile recovery environment. It buys IT time to remove the trigger files properly and potentially the infected data as well. The sterile recovery environment needs to deliver production-class performance and reliability to get the time IT needs to inspect the production environment thoroughly. The sterile environment also needs to protect itself with immutability. The need for a sterile recovery environment is new because of ransomware but is helpful for recovery from many other outages. To learn more, read our latest Whitepaper,” What is Standby Storage.”

Ransomware Recovery Requirement 5: Affordability

The ransomware recovery strategy has to be affordable. Part of a 360° Ransomware Recovery Strategy means that all applications and data are protected the same way. Today, IT has to piecemeal this type of strategy together using solutions from multiple vendors, which significantly raises the purchase, implementation, and operational costs. The fragmentation also forces IT to either pick which component of the strategy to implement, which leaves open attack vectors, or pick only a few data sets to protect, which means the rest of the environment is still exposed.

It should also intelligently leverage flash and high-density (20TB) hard disk drives to keep costs low. Using a hard disk tier means the solution needs to rapidly recover from a drive failure because RAID Rebuild times are more relevant than ever.

Delivering a 360° Ransomware Recovery Solution

Implementing a 360° Ransomware Recovery Strategy means IT needs to look for a single solution that can provide cost-effective backup storage while meeting the five previously listed requirements. The solution needs a small flash tier to handle the hourly ingest of backup data from dozens, if not hundreds of virtual machines, application servers, and NAS systems. The flash tier should be large enough to support recovery and enable customers to use backup-server applications with “instant recovery” features to exploit that capability fully.

A 360° Ransomware solution should also support the native restoration of data and become Standby Storage. It should provide native sterile volumes mountable via any storage protocol and enterprise-class performance and availability. Standby Storage is valuable for more than just ransomware recovery by also providing a recovery area during a storage controller failure or storage software failure.

Only StorONE’s S1:Backup delivers all of these capabilities in a single, cost-effective solution that is easy to implement and operate. To learn more, watch our on-demand webinar “Beat Ransomware with Standby Storage.”

Beat Ransomware with Standby Storage

January 11, 2022 by Drate Berry