Whether the method is old-fashioned backup and restoration or modern cloud-based disaster recovery, the need for archiving the work of the enterprise remains vital.
It can't escape those of us who appreciate the ironies of life that, while magnetic tape started vanishing in the '80s when the eight-track music tape lost popularity, tape has started disappearing from the computer industry in earnest just since the turn of the millennium.
It's ironic, of course, because the computer industry has for so long been culturally emblematic of technological progress. Yet, say the word "backup" to a computer person and most of them, if only for a second, will likely visualize a mag tape reel.
While there remain even today enterprises that back up systems to tape because they don't want to mess with what works, the past decade has been one of both significant technological improvement and cost-per-megabyte cost reductions in backup and recovery operations.
But for every traditionalist who brags that sticking with tape until now lets them upgrade to cloud-based high-availability overnight, there are hundreds who have embraced the advent of new technology in the pedestrian but necessary task of archiving our work. That's happened because the financial benefit of not having to pay for upgrading technology for this function has been outstripped by the opportunity cost of failing to do so.
Backup Windows the Size of Postal Boxes
"Data loss and extended downtime of any kind has become unacceptable for most modern organizations in the 21st century," notes Simon O'Sullivan, senior vice president at Maxava, a major high-availability vendor in the IBM i market. "The cost of real-time backup solutions, bandwidth, and related hardware has decreased dramatically, which in turn has increased the demand."
At the same time, enterprises that must operate 24x7, or close to it, have lost the ability to handle backup chores on a night or swing shift. Therefore, backup functions have to be integrated into regular processing times.
"Downtime is no longer tolerated by customers, and regulators are requiring real-time data replication," adds Robert Seal, owner of iSam Blue, a company that specializes in hardware-based HA.
The Tale of the Tape
There are essentially four technology types related to backup and recovery. Traditional backup is writing files to mag tape or disk that is stored somewhere nearby and used to recover anything lost on a short-term basis. Disaster Recovery (DR) writes files to disk or other media that's stored at a remote location; it's designed primarily to handle restoration of complete systems in the event of a calamity sufficiently dire to cause major data or server losses. High Availability (HA) sends images of whole systems to parallel operating servers, to which operations can be transferred in a short time period (and hopefully transparently to users) if primary systems become incapacitated or otherwise unavailable. Vaulting is an intermediate solution in which journal entries are sent to a storage device, to be added to a separate base backup to provide DR services as needed.
Although primarily offered via software products in the past, all of these technology types are increasingly being provided by hardware solutions as well, in addition to companies offering Internet-based services of all four types. All must be included in an assessment of the backup and recovery product and service market because the four types have become so closely related in strategic function for virtually all enterprises.
Two metrics play a role in evaluating the efficacy of such services. Recovery Time Objective (RTO) is the amount of time it takes to restore a particular level of service to users whose access to computing resources has been disrupted. This can obviously vary widely, depending on the level of service demanded, and includes time spent trying to fix a problem simply, the recovery processes themselves, and testing to be sure the recovery processes have worked. Recovery Point Objective (RPO) is a fixed time period that is considered tolerable for the RTO to take before there is an impact on the affected business itself. Clearly, this is affected by the type of business an enterprise engages in and customer and user demand for each IT service lost, which creates a limiting factor for any DR plan that affects what methods are used to provide DR.
Picking a backup strategy must take into account these metrics, and other features are popular as well. Seal summarizes as follows: "DR customers want to save to disk, have the process automated, consolidate backups to one device for all their servers, and automate sending data to their own offsite storage. HA customers want fair pricing and lower costs, a system that's easy to use, and better autonomics."
The Hazards of Making No Choice at All
Procrastinating on picking a strategy is becoming increasingly risky.
"2011 was the most expensive year in history for natural disasters," points out O'Sullivan. "IT managers are under pressure to back up more data faster with zero data loss." He adds, "Ten years ago, taking a daily tape backup plus a monthly full system save and either sending tapes offsite or putting tapes in a safe was considered normal practice. Businesses now demand a better RTO and RPO, which has led to the increase in real-time solutions that capture and send transactions to a remote hot server."
The problem is likely to increase over time. "More companies are going to fail because they lose even a few hours of data," Seal predicts. "As this happens, companies are going to understand that losing any data is not acceptable."
Despite the expense, handling backups, DR, or even full-blown HA via the cloud is appealing because of the potential simplicity for in-house personnel and overall efficiency, despite the traditional objection of loss of control over physical backup media. Of course, the perception of HA remains one of being an expensive alternative that has to be justifiable.
"Cloud backups are a game changer," emphasizes O'Sullivan, "because they eliminate the need to FedEx tapes between locations and they allow very quick recovery times. Local backups are great if a single file or library needs to be restored (and they probably always will be), but DR/HA in the cloud is different. It's becoming less expensive, it's faster and less intrusive on the business, it's an operational expense, and security is handled by the managed service provider. Locally stored backup media as part of a DR plan is a bad idea, and it's time has passed."
"Customers are looking to simplify and cut costs. Having 'tape' images automatically sent off site does this nicely," Seal opines. "Control, access, and security are all issues, but with reduced budgets and increased demands in other areas, I believe [cloud-based backups] are going to grow until HA is seen as being more affordable."
The Backup/Recovery/DR/HA Market for IBM i Today
Below are representative listings for major backup and recovery, DR, and HA product and service offerings for the IBM i, divided into general backup products and services, and those specifically aimed at the DR and HA markets. These listings cover such products for general systems use but not, for example, database-specific or database-only replication tools, report- or spooled-file archiving products, and file-transfer products, even though they are of some use in archiving data. Please note that the brief descriptions here don't cover all the features each offering provides. You should consult the links provided for each product and contact the associated vendors for a more complete idea of what each product's or service's capabilities include.
And as always when looking for products or services, be sure to check the MC Press Online Buyer's Guide.
Backup and Recovery Products and Services for IBM i
ASG Software Solutions
Atempo Time Navigator is a backup and recovery solution for saving and restoring files from any point in time from any backup media for Linux, UNIX, and Windows servers.
Asigra Cloud Backup is designed to provide backup and recovery services for public, private, and hybrid cloud environments. It can be purchased either as software for onsite personnel to use or as a SaaS service offering managed by Asigra.
Champion Solutions Group
Champion's Cloud Storage Strategies provides cloud-based backup and recovery, long-term archiving, and disaster-recovery services.
Server Virtualization services include backup/recovery/data-sharing with virtual servers, server consolidation, and disaster recovery.
CPS Technology Solutions
Unitrends is a family of disk-to-disk backup appliances for Linux and Windows systems.
Data Storage Corporation
SafeData/Cloud offers cloud-based "capacity-on-demand" backup services for IBM i, AIX, and Windows applications.
The SafeData/Vault service supports centralized backup and recovery infrastructures featuring offsite storage and recovery.
Avamar is backup and data deduplication software for IBM i and other servers running AIX and Linux. It emphasizes full daily backups and one-step recovery processes.
EVault is disk-to-disk data backup and recovery software for IBM i servers running i5/OS, AIX, or Linux, as well as other platforms. It builds backups dynamically to avoid recovery from incremental backups and encrypts backup data during transmission.
EVault also offers its backup software as part of a cloud-based backup service for a wide range of computer platforms.
Robot/SAVE includes automatic backup and recovery, backup-tape encryption, Lotus Domino online backup, and other features.
DBMoto Cloud Edition provides cloud-based data replication for moving or sharing data with remote systems.
Hosting.com incorporates cloud- and EMC-based backup and recovery services.
The CBU with PowerHA program offers secondary servers that can form the basis for HA and DR services for production machines.
Blue Cloud Services provides cloud-based backup and recovery and HA for IBM i and open-systems servers.
LXI Enterprise Storage
The iSeries Open Client offering lets enterprises control backups of multiple open-systems servers from a single console.
Information Lifecycle Management includes backup and recovery management, tape management, vault management, Lotus Notes/Domino hot backup, and tape encryption products and services for IBM i, AIX, and Linux servers, as well as servers using other operating systems.
Pinnacle Business Systems
TRAC 400 software handles backup and recovery and media-management operations across multiple servers and interfaces with virtually all job-scheduling utilities.
SonicWALL's Backup and Disaster Recovery Platform is a family of backup and recovery appliances for servers and desktops running Linux and Windows.
Antares is a backup storage appliance for IBM i, HP, and open-systems servers running Linux, UNIX, and Windows.
Eclipse is a backup storage appliance for IBM i, Power Systems, PureSystems, and HP OpenVMS servers.
Storage Director is a hardware device that consolidates, optimizes, and encrypts backup storage operations to disk from a variety of other sources.
High-Availability/Disaster-Recovery Products and Services for IBM i
Bug Busters Software Engineering
RSF-HA replicates data, libraries, user profiles, system values, network attributes, and authorization lists to a hot backup machine and automatically creates journals needed to create an HA system.
Business Continuity Specialists
The Continuity and Security Suite for IBM i includes Quick-EDD/HA, an HA application that provides data synchronization, dynamic monitoring between source and target systems, and optional backup to multiple machines. It also includes Quick-EDD/DR, data-replication software for DB2 databases on i servers.
Data Storage Corporation
The iSeries Disaster Recovery program is a service that includes backup and recovery testing, disk-based data backup, and data encryption.
Data Storage's iSeries High Availability is a hosted-service solution for HA that features disk-based storage, recovery in 15 minutes or less, and support for AIX, Linux, and Windows as well as i5/OS.
Virtual Recovery is a hardware-free virtual backup and recovery service for IBM i that backs up data to a remote data center and restores it on demand via a virtual private network protected by password access and data encryption.
Advanced Copy Services for Power HA uses FlashCopy storage-volume replication and Independent Auxiliary Storage Pools (IASPs) to provide HA and DR services.
Power HA SystemMirror for i is a hardware solution for automating high-availability and disaster-recovery operations. It offers a clustering solution that requires no downtime to operate and HA/DR protection with simplified setup.
The iSB/RSF-HA offering provides simplified HA and disaster-recovery services for IBM i servers.
Maxava HA Datastream runs on backup servers to provide a one-way, data-only replication engine.
Maxava HA Enterprise+ helps automate and manage HA role swaps between multiple servers by featuring real-time replication of data and applications, the maxview Manager utility for monitoring and controlling HA activities, and a command-scripting function.
Maxava HA SMB, aimed at SMB enterprises, facilitates replication of data and objects, offers on-demand replication of data queues and data areas, and provides a unified monitoring capability.
maxView is a suite of monitoring tools for HA environments. It includes maxView Manager (for interacting with Maxava HA Enterprise+), maxView Monitor (for oversight and interaction with Maxava HA SMB), and maxView Lite (for viewing IBM i system status from mobile devices).
Shield Advanced Solutions
HA4i is an HA solution for IBM i that uses Remote Journal functions to provide a simpler, easy-to-install, and uncomplicated-to-manage alternative to larger HA systems. HA41 includes a PHP-based user interface that lets users run everything from a single console.
SunGard Availability Services includes cloud-based HA, managed infrastructure-as-a-service (IaaS), consulting, and business-continuity software.
Vision's Double-Take Availability provides HA services for servers running AIX and Linux, including IBM i. Providing failover protection for those operating systems, the product replicates to remote DR sites via standard IP networks.
Based on remote journaling, iTera Availability maintains a complete backup of production and other designated systems and switches users to the backup system on demand.
MIMIX Availability replicates applications, data, and system values in real time from a production server to a backup server. Offered in three versions with accumulating features for enterprises of various sizes, the product ensures that backup systems are ready to run as needed.
MIMIX Global is designed to handle multiple-node HA installations. Equipped with a browser interface and available in three editions of varying capabilities, MIMIX Global provides automated help for maintaining backup systems, alerting personnel to problems requiring switchovers, and managing transitions to backup systems.
MIMIX RecoverNow is a solution that automates disaster-recovery operations for IBM i systems. It captures activities between tape backups, simplifies journal management, and guides operations personnel step-by-step through any necessary recovery procedures.
OMS/ODS, designed for larger enterprises, maintains exact or near-real-time images of multiple systems. Images include all applications, data, and system objects. Its multi-CPU design can replicate environments of any complexity and keep them available for failover at any time.