Automated Data Libraries Can Help To Manage Your Data In Y2K.
Fear: the overriding reason for action or inaction. Uncertainty: the overriding reason for change or a lack of change. Doubt: the overriding reason for a lack of trust or for trust. These concepts are commonly referred to as FUD and used to prevent change from one vendor to another. With two opposite views of the same concept, confusion and chaos may reign.
Y2K: a reason for action; a reason for change; a reason for lack of trust. FUD is alive and well in corporate America. No longer can one trust any one individual to perform without aggressive follow-up and checking on the results. No longer can one delay changing the data environment due to uncertainty since the price of failure is so high. No longer can one delay action in protecting corporate data and "blame" failure on someone else.
The Y2K problem is not unknown to most corporations. Corporate America has been trying to do something about it. But is any one company in control of its own data environment? Or, are almost all companies dependent upon serious amounts of data being "interlocked" with another company's (suppliers, customers, and government) data?
Can a company trust the competence and the integrity of its employees? A certain amount of prevention of disaster may be the real issue rather than "lambasting" employees for any Y2K reason.
Company Y2K Actions
Many companies have spent many dollars on re-doing, testing and re-implementing Y2K-affected software. But, did they find all of the problems? Fear, Uncertainty, Doubt.
Should a company freeze the introduction of new software systems after a certain date, say July 1, 1999, in anticipation of Y2K? Should the old systems be updated after July 1? Can storage issues be ignored for six months or more until after Y2K settles down?
How does a company insulate itself from its suppliers? From its customers? Shutting down operations temporarily seems "anti" to the mission of the business. How much data needs to be duplicated?
Even scarier, a substantial number of international countries and corporations do not have the resources to deal effectively with the Y2K problem quickly, or before the occurrence. The U.S. and Canada face a very real risk in dealing with the international aspects of their business.
Several techniques are being tried, but the issue very quickly becomes how many sets of data need to be kept for recovery and experimentation purposes. When with a problem how quickly can a company recover? Recovery implies that data is not "missing"!
A company must have multiple copies of data to protect itself. The issue is how many copies and how to effectively double the recording bandwidth of data. But how does a company manage the suddenly increased amounts of data? Software? Faster drives for increased bandwidth? Sharing of storage resources for bandwidth and capacity efficiency? High performance RAID systems with offline mirroring and backup? Automated data libraries for online storage?
The Library Supplier Challenge
The suppliers of automated data libraries (ADLs) are in a unique position to provide the solutions necessary for a company to ameliorate the effects or FUD of the Y2K problem.
The basic ADL provides the cost-effective capacity and the hardware capability to have the data managed online. The user, or some other "expert or consultant", needs to determine the amount of data, the data flow parameters and the configuration of the data paths. All vendors provide good to excellent ADLs--the user only needs to decide the capacity, the technology type and the number of drives with each ADL.
It's simple--but only for the home user who needs only one tape drive.
The complexity comes in understanding the data flow parameters and the required access times for a variety of online applications. The discipline to implement a variety of data duplication protocols within the enterprise is also required.
Several companies provide data mirroring hardware configurations and several software products are available to provide data mirroring from a software package. With data mirroring, a duplicate copy of data is created with one I/O command in almost the same time as a single data set can be created. Data mirroring can be implemented easily into automated data libraries with a minimal effort.
There are several SAN data gateways coming to market that allow for tremendous user benefits at a minimal cost. These data gateways, from companies such as IBM, StorageTek and Pathlight Technology plus others, allow Fibre Channel connections into the data gateway--yielding high-speed data transmission--and SCSI connections to attach the existing storage peripherals, such as ADLs. All attachments are transparent to the applications and the host servers. With "zoning", individual servers can have specific peripherals or parts of ADLs allocated to only themselves. With these newer implementations, one does not need a plethora of pieces, e.g., routers, converters, etc.
With Y2K, a company either deals with the Fear, Uncertainty and Doubt or goes into an irreversible decline. Utilizing ADLs as a significant part of the Y2K solution only makes sense. ADLs will be with us for the foreseeable future as all data goes online and is managed online throughout the enterprise.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technology Information|
|Author:||Elizer, Lee H.|
|Publication:||Computer Technology Review|
|Date:||Jul 1, 1999|
|Previous Article:||DataDirect Networks Hoping For A Fit With Its SANware.|
|Next Article:||Ultra160/m Will Become The Preferred SCSI Implementation.|
|Y2K: The Demand for Evidentiary Records.|
|Key software issues to consider when purchasing a tape library.|