Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Gorilla and King Portfolio Candidates -- Ignore unavailable to you. Want to Upgrade?

To: gdichaz who wrote (26332)	6/14/2000 10:43:00 AM
From: DownSouth	Read Replies (1) \| Respond to of 54805

NTAP customers do connect filers over long distances for the purpose of SNAPMIRROR. SNAPMIRROR allows the automatic backup of a filer's volume(s) (and subsequent "incremental" updates) to be requested by and received by a target filer. The request and transfer are via TCP. It is up to the customer to provide the appropriate bandwidth to keep up with the SNAPMIRROR schedule. Here is a good discussion: netapp.com Another connection between filers that most new customers are implementing is a pair of clustered failover filer partners. This connection is an FC connection, so, at this time, there is no LAN/WAN involved. Clustered failover configs allow two filers to serve as hot backup to one another. Both filers are fully active, but are monitoring one another's health and keeping copies of one anothers write cache (NVRAM). When one filer fails, the other finishes the failed filers cached writes and takes over that filers IP addresses, transparent to the users. The partner accesses the failed partner's RAID arrays through the dual channel FC (FC-AL) that both are sharing. Recovery of the failed filer results in a "fail back" to the recovered filer so that all is normal again, transparent to the users. Here is a good discussion of clustered failover: netapp.com This clustering with new FC switches and IP interfaces from BRCD/CSCO will lead to new capabilities as hinted at here: It is conceivable that in the future users may access their data through a sort of "global" file system comprised of multiple servers in different geographical locations, each replicating portions of the others data. Access to the active file system would always be to the physically closest server, but access to older versions of the data (backup/archive restores) may be to a Snapshot on the local server or on a geographically remote server. With the price per megabyte of hard disk storage having been reduced by a factor of ten or more in the past few years, this is a much more realistic scenario than it might first appear. This is from the first article that I cited above.