Redundancy in Data Storage: Part 2: Geographical Replication

This is a followup to my previous article, Redundancy in Data Storage: Part 1: RAID Levels, where I discussed various site-local data redundancy technologies. Here, I will attempt to detail many of the choices available to provide redundancy beyond the data center that organizations use to solve disaster recovery, business continuity, and continuity of operations (COOP).

It’s obvious that site-local redundancy isn’t enough for critical applications. The threat of natural disasters is always looming, regional power outages occur, building electrical and mechanical systems fail, and backhoes seem to hate fiber optic cable. Enterprises therefore attempt to use geographic redundancy to ensure that even when these things happen critical applications and data remain available. At the heart of making an application geographically redundant is making sure the application’s data resides in more than one geographical location. There are a number of technology and architectural choices that can be used to achieve this geographical replication of data. Often these solutions will be evaluated in terms of cost, RTO (recovery time objective), and RPO (recovery point objective), as I outlined in Disaster Recovery and the Cloud.

One obvious place to build redundancy is at the storage area network level. There are a variety of technologies available to replicate SAN volumes between geographic locations. Synchronous replication tightly couples the primary and backup sites and does not return success to the storage controller until a write completes in both locations, providing a zero RPO. However, synchronous replication requires very fast network connections and requires that the backup site be located very close to the primary location because otherwise latency will severely reduce storage performance. To allow the sites to be further apart, asynchronous replication can be used where the changes are streamed to the backup site but completion of the I/O is signalled before receiving an acknowledgement. Finally, point-in-time replication generates many snapshots of the storage and sends the delta between each snapshot.

All of these SAN replication approaches are bandwidth intensive. Applications make many changes to the disk as part of their ordinary functioning and these changes are almost certainly not encoded in a dense fashion that allows them to efficiently cross networks. An application might make small updates to the same disk block many times in short order and all of these changes would have to be sent across the network in asynchronous or synchronous replication. Point in time replication lowers this overhead a small amount (because redundant changes between snapshots are not sent) at the cost of worse RPO.

Redundancy can also be implemented through database replication. Just as in SAN replication, synchronous, asynchronous, and snapshot-based techniques are available. Many of the same tradeoffs apply, although generally database changes can be sent more efficiently across a WAN. Unfortunately, effectively using database replication to provide geographic redundancy is difficult. For one, database replication can only stand on its own if all of the critical application data resides within the database. This is often not the case. Moreover, sophisticated database deployments involving data partitioning, federation, and integration often greatly complicate replication to the point that effective configuration becomes prohibitive.

Finally, the application itself can handle data redundancy. Often the highest end applications (for instance financial, logistics, and reservation systems) require the federation of data at the application level. This allows extreme top-end performance to be reached and also allows compliance with various types of data jurisdiction requirements (for instance, national directives requiring customer identifiable information to remain in the country of origin.) Unfortunately, this is very difficult and error prone.

Data redundancy is only one piece of the business continuity problem. Applications require other infrastructure to run, such as the network and application servers. Some organizations are using virtualized approaches here with some success to build geographically redundant architectures. Others rely on configuration management technologies to ensure that the disaster recovery sites remain synchronized and ready to handle workload. Another important point to consider is how to handle moving the active instance of the application to the backup site, and also how to re-establish redundancy after a failure and move applications back to the primary. Any approach to provide geographic redundancy must be designed carefully and continually tested well, because today’s complicated application architectures provide too many opportunities for mistakes to be made in provisioning redundancy.

These replication techniques still require the site-local mechanisms like RAID discussed in part 1, because otherwise the facilities involved would be far too unreliable, and also require significant investments in network links, replication technologies, and personnel effort. Also, for the most part, these technologies require the duplication of infrastructure for disaster recovery purposes. In my forthcoming part 3, I will discuss emerging approaches in cloud architectures that unify redundancy mechanisms and significantly simplify the effort involved in implementing resilient business systems.

About the Author

Michael Lyle (@MPLyle) is CTO and co-founder of Translattice, and is responsible for the company’s strategic technical direction.  He is a recognized leader in developing distributed systems technologies and has extensive experience in datacenter and information technology operations.

GD Star Rating

Redundancy in Data Storage: Part 1: RAID Levels

I recently read Joe Onisick’s piece, “Have We Taken Data Redundancy Too Far?”  I think Joe raises a good point, and this is a natural topic to dissect in detail after my previous article about cloud disaster recovery and business continuity.  I, too, am concerned by the variety of data redundancy architectures used in enterprise deployments and the duplication of redundancy on top of redundancy that often results.  In a series of articles beginning here, I will focus on architectural specifics of how data is stored, the performance implications of different storage techniques, and likely consequences to data availability and risk of data loss.

The first technology that comes to mind for most people when thinking of data redundancy is RAID, which stands for Redundant Array of Independent Drives.  There are a number of different RAID technologies, but here I will discuss just a few.  The first is mirroring, or RAID-1, which is generally employed with pairs of drives.  Each drive in a RAID-1 set contains the exact same information.  Mirroring generally provides double the random access read performance of a single disk, while providing approximately the same sequential read performance and write performance.  The resulting disk capacity is the capacity of a single drive.  In other words, half the disk capacity is sacrificed.

RAID-1, or Mirroring

RAID-1, or Mirroring;
Courtesy Colin M. L. Burnett

A useful figure of merit for data redundancy architectures is MTTDL, or Mean Time To Data Loss, which can be calculated for a given storage technology using the underlying MTBF, Mean Time Between Failures, and MTTR, Mean Time To Repair/Restore redundancy.  All “mean time” metrics really specify an average rate over an operating lifetime; in other words, if the MTTDL of an architecture is 20 years, there is a 1/20 = approximately 5% chance in any given year of suffering data loss.  Similarly, MTBF specifies the rate of underlying failures.   MTTDL includes only failures in the storage architecture itself, and not the risk of a user or application corrupting data.

For a two-drive mirror set, the classical calculation is:

This is a common reason to have hot-spares in drive arrays; allowing an automatic rebuild significantly reduces MTTR, which would appear to also significantly increase MTTDL.  However…

While hard drive manufacturers claim very large MTBFs, studies such as this one have consistently found numbers closer to 100,000 hours.  If recovery/rebuilding the array takes 12 hours, the MTTDL would be very large, implying an annual risk of data loss of less than 1 in 95,000.  Things don’t work this well in the real world, for two primary reasons:

  • The optimistic assumption that the risk of drive failure for two drives in an array is uncorrelated.  Because disks in an array were likely sourced at the same time and have experienced similar loading, vibration, and temperature over their working life, they are more likely to fail at the same time.  Also, some failure modes have a risk of simultaneously eliminating both disks, such as a facility fire or a hardware failure in the enclosure or disk controller operating the disks.
  • It is also assumed that the repair will successfully restore redundancy if a further drive failure doesn’t occur.  Unfortunately, a mistake may happen if personnel are involved in the rebuild.  Also, the still-functioning drive is under heavy load during recovery and may experience an increased risk of failure.  But perhaps the most important factor is that as capacities have increased, the Unrecoverable Read Error rate, or URE, has become significant.  Even without a failure of the drive mechanism, drives will permanently lose blocks of data at this specified (very low) rate, which generally varies between 1 error per 1014 bits read for low-end SATA drives to 1 per 1016 for enterprise drives.  Assuming that the drives in the mirror are 2 TB low-end SATA drives, and there is no risk of a rebuild failure other than by unrecoverable read errors, the rebuild failure rate is 17%.
RAID 1+0: Mirroring and Striping

RAID 1+0: Mirroring and Striping;
Courtesy MovGP

With the latter in mind, the MTTDL becomes:

When the rebuild failure rate is very large compared to 1/MTBF:

In this case, MTTDL is approximately 587,000 hours, or a 1 in 67 risk of losing data per year.

RAID-1 can be extended to many drives with RAID-1+0, where data is striped across many mirrors.  In this case, capacity and often performance scales linearly with the number of stripes.  Unfortunately, so does failure rate.  When one moves to RAID-1+0, the MTTDL can be determined by dividing the above by the number of stripes.  A ten drive (five stripes of two-disk mirrors) RAID-1+0 set of the above drives would have a 15% chance of losing data in a year (again without considering correlation in failures.)  This is worse than the failure rate of a single drive.


RAID-5 and RAID-6;
Courtesy Colin M. L. Burnett

Because of the amount of storage required for redundancy in RAID-1, it is typically only used for small arrays or applications where data availability and performance are critical.  RAID levels using parity are widely used to trade-off some performance for additional storage capacity.

RAID-5 stripes blocks across a number of disks in the array (minimum 3, but generally 4 or more), storing parity blocks that allow one drive to be lost without losing data.  RAID-6 works similarly (with more complicated parity math and more storage dedicated to redundancy) but allows up to two drives to be lost.  Generally, when a drive fails in a RAID-5 or RAID-6 environment, the entire array must be reread to restore redundancy (during this time, application performance usually suffers.)

While SAN vendors have attempted to improve performance for parity RAID environments, significant penalties remain.  Sequential writes can be very fast, but random writes generally entail reading neighboring information to recalculate parity.  This burden can be partially eased by remapping the storage/parity locations of data using indirection.

For RAID-5, the MTTDL is as follows:

Again, when the RFR is large compared to 1/MTBF, the rate of double complete drive failure can be ignored:

However, here RFR is much larger as it is calculated over the entire capacity of the array.  For example, achieving an equivalent capacity to the above ten-drive RAID-1+0 set would require 6 drives with RAID-5.  The RFR here would be over 80%, yielding little benefit from redundancy, and the array would have a 63% chance of failing in a year.

Properly calculating the RAID-6 MTTDL requires either Markov chains or very long series expansions, and there is significant difference in rebuild logic between vendors.  However, it can be estimated, when RFR is relatively large, and an unrecoverable read error causes the array to entirely abandon using that disk for rebuild, as:

Evaluating an equivalent, 7-drive RAID-6 array yields an MTTDL of approximately 100,000 hours, or a 1 in 11 chance of array loss per year.

The key things I note about RAID are:

  • The odds of data loss are improved, but not wonderful, even under favorable assumptions.
  • Achieving high MTTDL with RAID requires the use of enterprise drives (which have a lower unrecoverable error rate).
  • RAID only protects against independent failures.  Additional redundancy is needed to protect against correlated failures (a natural disaster, a cabinet or backplane failure, or significant covariance in disk failure rates.)
  • RAID only provides protection of the data written to the disk.  If the application, users, or administrators corrupt data, RAID mechanisms will happily preserve that corrupted data.  Therefore, additional redundancy mechanisms are required to protect against these scenarios.

Because of these factors, additional redundancy is required in conventional application deployments, which I will cover in subsequent articles in this series.

Images in this article created by MovGP (RAID-1+0, public domain) and Colin M. L. Burnett (all others, CC-SA) from Wikipedia.

This series is continued in Redundancy in Data Storage: Part 2: Geographical Replication.

About the Author

Michael Lyle (@MPLyle) is CTO and co-founder of Translattice, and is responsible for the company’s strategic technical direction.  He is a recognized leader in developing new technologies and has extensive experience in datacenter operations and distributed systems.

GD Star Rating

Disaster Recovery and the Cloud

It goes without saying that modern business relies on information technology.  As a result, it is essential that operations personnel consider the business impact of outages and plan accordingly.  As an illustration, Virgin Blue recently experienced a twenty-hour outage in its reservation system that resulted in losses of up to $20 million dollars.  The cloud provides both considerable opportunities and significant challenges relating to disaster recovery.

In general, organizations must currently build multiple levels of redundancy into their systems to reach high-availability targets and to protect themselves from catastrophic outages during a natural or man-made disaster.  A disaster recovery strategy requires that data and critical application infrastructure be duplicated at a separate location, away from the primary datacenter.  Cutting over to a disaster recovery site is usually not instantaneous and redundancy is often lost during the contingency operating plan.  For this reason, site-local redundancy mechanisms – such as high availability network systems, failover for portions of the application stack, and SAN-level redundancy are also required to achieve availability goals.  Public clouds often further complicate disaster recovery planning, as the organization’s critical systems may now be spread across their own infrastructure and a multitude of outside vendors, each with their own data model and recovery practices.

Business requirements and application criticality should guide the approach chosen for business continuity.  Consider the concepts of RPO (Recovery Point Objective) and RTO (Recovery Time Objective). The RPO of a system is the specified amount of data that may be lost in the event of a failure, while the RTO of a system is the amount of time that it will take to bring the system back online after a failure.  In general, site-local mechanisms will provide near-instantaneous RPO and RTO, while disaster recovery systems often will have an RPO of several hours or days of information, and an RTO measured in tens of minutes. Through increasingly sophisticated (and costly) infrastructures, these times can be reduced but not entirely eliminated.

Timeline illustrating concepts of RPO and RTO

Illustration of RTO and RPO in a backup system

Dedicated redundancy infrastructure, both site-local and for disaster recovery purposes, must be regularly tested.  Additionally, it is essential to ensure that the disaster recovery environment is compatible with the existing infrastructure and capable of running the critical application.  This is an area where change management procedures are important, to ensure that critical changes to the production infrastructure are made in the standby environment as well.  Otherwise, the standby environment may not be able to correctly run the application when the disaster recovery plan is activated.

The primary factor that determines RTO and RPO is the approach used to move data to the contingency site.  The easiest and lowest cost approach is tape backup.  In this case, the RPO is the time between successive backups moved off-site (perhaps a week or more) and the RTO is the amount of time necessary to retrieve the backups, restore the backups, and activate the contingency site.  This may be a significant amount of time, especially if personnel are not readily available during the disaster scenario.  Alternatively, a hot contingency site may be maintained, and database log-shipping or volume snapshotting/replication can be used to send business data to the secondary site.  These systems are costly, but readily attain an RTO of under an hour, and an RPO of perhaps one day.  With substantial investment and complexity, RPO can even be reduced to the range of minutes.  However, organizations have often been surprised to find that the infrastructure doesn’t work when it is called upon, often because of the complexity of the infrastructure and the difficulties involved in testing a standby site.

When procuring IaaS (Infrastructure as a Service) or SaaS (Software as a Service), it is essential for the organization to perform due diligence regarding what disaster recovery mechanisms the service vendor uses. The stakes are too high to trust service level agreements alone (in the case of a catastrophic failure during a disaster, will the vendor be solvent and will the compensation received be sufficient to compensate for business losses?).

Disaster Recovery as a Service, or DRaaS, is an emerging category for organizations that wish to control their own infrastructure but not maintain the disaster recovery systems themselves.  With a DRaaS offering, an IT organization does not directly build a contingency site, but instead relies on a vendor to do so on a dedicated or utility computing infrastructure.  The cloud’s advantages in elasticity and cost-reduction are significant benefits in a disaster recovery scenario, and service offerings allow organizations to outsource portions of contingency planning to vendors with expertise in the area.  However, many of the complexities remain and it is essential to perform the due diligence to ensure that the contingency plan will work and provide a sufficient level of service if called upon.

Finally, there are emerging technologies that combine site-local redundancy and disaster recovery into a unified system.  For example, distributed synchronous multi-master databases allow an application to be spread across multiple locations, including cloud availability zones, with the application active and processing transactions in all of them.  A specified portion of the system can be lost without any downtime or recovery effort.  These emerging systems offer the prospect of dramatically reducing costs and minimizing the risk of contingency sites not functioning properly.

About the Author

Michael Lyle (@MPLyle) is CTO and co-founder of Translattice, and is responsible for the company’s strategic technical direction.  He is a recognized leader in developing new technologies and has extensive experience in datacenter operations and distributed systems.

GD Star Rating