toasters

toasters@lists.teaparty.net

13531 discussions

RE: Looks like InfoLibria got the last laugh at the cache bakeoff .
by Mohler, Jeff 07 Apr '99

07 Apr '99

*heh* Cute. Of course, my management caught wind of this too..and lacking NetApp information, I may not be able to change the tide for our backbone. > -----Original Message----- > From: Jaye Mathisen [SMTP:mrcpu@internetcds.com] > Sent: Wednesday, April 07, 1999 2:16 AM > To: toasters(a)mathworks.com > Subject: Looks like InfoLibria got the last laugh at the cache > bakeoff. > > > > > http://www.infolibria.com/press/040599.htm > > > > Already had a netcache appliance, didn't buy one. Looks like time to give > infolibria a call, since they're saying a 5-9 month payback, and see if > they want to toss me a loaner for eval... > > >

1 0

Looks like InfoLibria got the last laugh at the cache bakeoff.
by Jaye Mathisen 07 Apr '99

07 Apr '99

http://www.infolibria.com/press/040599.htm Already had a netcache appliance, didn't buy one. Looks like time to give infolibria a call, since they're saying a 5-9 month payback, and see if they want to toss me a loaner for eval...

1 0

RE: F740 cluster
by Smith, Gil 06 Apr '99

06 Apr '99

> -----Original Message----- > From: beepy(a)netapp.com [mailto:beepy@netapp.com] > I was going to slip "The Downward Spiral" onto our corporate phone > system background music about four years ago - but far be it from > me to dictate my musical tastes on everyone else. Which is why most people (at least I did when I was a UNIX admin) prefer silence with maybe an auto attendant that will let you know the approx. wait time and how many people are ahead of you in the queue. At least then you can decide whether or not to remain on hold -- possibly offered the option to leave a message. --Gil

3 2

RE: IRCACHE Web Cache Bakeoff?
by Mohler, Jeff 06 Apr '99

06 Apr '99

> NetApp is pleased with its performance in the NLANR bake-off and are > sharing our results with customers and potential customers under NDA. > We have decided not to publicly release our results at this time > because of the late withdrawal from the bake-off by the other major > caching vendors (Cisco and Inktomi). Network Appliance does not want > to disclose detailed performance numbers to our competitors without > having access to their results. Based on past experience we believe > that competitors are likely to compare apples to oranges and > selectively quote our results as it suits their requirements. --- Let em cut their own throats. When someone learns that their stuff doesnt measure up...THEIR loss. It's distressing internally for me at Willams, and personally depressing to see the "Theyre not, so we're not" spin.

1 0

"df" on 5.3
by Graham C. Knight 06 Apr '99

06 Apr '99

Does anyone know why the "df" output on 5.3 was changed? It no longer lists the snap reserve usage. It does list the snapshot reserve if nosnapdir is "off" - but our default is "on". 5.1 and 5.2 listed it regardless of the option's state. This breaks all my snapshot monitoring scripts :-( Why does netapp fiddle with the output of commands like this? Graham

1 0

Re: Muzak (was F740 cluster)
by dnb＠ccs.neu.edu 06 Apr '99

06 Apr '99

"Skinner, Alistair" <alistair.skinner(a)netapp.com> writes: > We knew people were having trouble with Kenny G, so we changed > recently to a longer sample of different music. However, it sounds > like that isn't cutting it either, and so we are purchasing an audio > feed card for our Aspect phone system. This will allow us to have a > live feed instead of a loop. Great! Does this mean we get to hear David Hitz's Greatest Hitz? Peace, dNb P.S. Ok, sorry. It's been one of those degraded mode days.

2 1

RE: NAC SAN
by Chen, Ray 06 Apr '99

06 Apr '99

Well, I wouldn't go so far as to call it "the" clarification of Dave's position. It's certainly *my* position and my take on your question "why focus on the protocol?" I'll let Dave decide whether or not he wants to let it stand or chime in with something different. Sorry if I didn't make that obvious in the original message. Ray > -----Original Message----- > From: Adams, Christian [mailto:Adams_Christian@emc.com] > Sent: Monday, April 05, 1999 2:32 PM > To: 'Chen, Ray'; Adams, Christian; Hitz, Dave; willh(a)infi.net > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > Subject: RE: NAC SAN > > > Hey Ray Chen - > > Thanks for the clarification of Dave Hitz's position. > > /Christian Adams > EMC > > > -----Original Message----- > > From: Chen, Ray [SMTP:ray.chen@netapp.com] > > Sent: Sunday, April 04, 1999 10:19 AM > > To: 'Adams, Christian'; Hitz, Dave; willh(a)infi.net > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > Subject: RE: NAC SAN > > > > Um, because the distinctive characteristic of SAN is that > > clients can write directly to the storage devices bypassing > > any server? This is possible only if a storage-level (e.g. > > block-level) protocol like SCSI runs on the wire. (Actually, > > I'm lying here. There are two ways of enabling this. One is > > to use a really simple protocol like SCSI. The other is to > > implement an industry-standard clustered filesystem with the > > disk drives/storage devices as an intelligent component of > > said filesystem. A very complex protocol whose viability > > and practicality is an issue in its own right. The rest of this > > mail assumes a block-level SAN protocol.) > > > > As folks have pointed out, SAN has its advantages and disadvantages. > > It eliminates the backplane of server (unless you have a > RAID controller > > in front of the drives :-) from the I/O path, eliminating a possible > > I/O bottleneck. But there is no server to protect the integrity > > of the data and even if there is, the protocol is so low-level in > > a SAN, the server has no effective way of detecting when a client > > has gone insane (maliciously or not) and is scribbling gibberish > > on the drives. So trust becomes a definite issue. > > > > Personally, I see data sharing using SANS as making sense > > in small workgroup environments and what I call "machine room > > clusters" -- a SAN in a machine room supporting a dedicated > > cluster. Typically you'd do the latter because scaling via > > a cluster is cheaper (and more available) than scaling via MP. > > But both are situations where you are willing to trust all the > > machines in the cluster because they are all firmly controlled by > > the group whose storage/data is being shared. > > > > Ray Chen > > rcc(a)netapp.com > > > > > -----Original Message----- > > > From: Adams, Christian [mailto:Adams_Christian@emc.com] > > > Sent: Tuesday, March 30, 1999 10:14 AM > > > To: 'Dave Hitz'; willh(a)infi.net > > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > > Subject: RE: NAC SAN > > > > > > > > > Hey Dave Hitz - > > > > > > Why do you focus on existing protocols in your discussion? > > > This indicates > > > that the introduction of Fibrechanel with either > encapsulated SCSI or > > > Ethernet is the critical event in the creation of SAN or NAS. > > > Further in > > > your discussion, you draw a parallel to switches and routers > > > which indicate > > > that you may be thinking in terms of shared versus segmented > > > data flows. > > > And yet in all of your discussion, you do not address the > issues of > > > management, integration, or value. > > > > > > Loved your EMC bit, > > > /Christian Adams > > > EMC > > > > > > > -----Original Message----- > > > > From: Dave Hitz [SMTP:hitz@netapp.com] > > > > Sent: Saturday, March 27, 1999 1:41 PM > > > > To: willh(a)infi.net > > > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > > > Subject: Re: NAC SAN > > > > > > > > > Tom's statement below (to a [NetApp] SAN when we build > > > one) brings up a > > > > > familiar argument between myself and my boss. My boss > > > contends that > > > > > NetApp is NOT capable of migrating to a "real SAN." > > > > > > > > I've been delinquent in my duty of writing a NAS and > SAN paper. I'm > > > > about half way done. Whem I'm done, I'll post it here, > probably in > > > > early form, so that you guys can critique it for me. > > > > > > > > Very briefly, I believe that NAS and SAN are different > but related > > > > technologies, and that the definitions are as follows: > > > > > > > > NAS (Network Attached Storage) > > > > Storage accessed over TCP/IP, using > industry standard file > > > > sharing protocols like NFS, HTTP and > Windows Networking. > > > > > > > > SAN (Storage Area Network) > > > > Storage accessed over a Fibre Channel > switching fabric, > > > > using encapsulated SCSI. > > > > > > > > I also think that both NAS and SAN are useful, and that -- > > > in the long > > > > run -- both will be critical components in any large data > > > > infrastructure. I think there's a strong analogy to the TCP/IP > > > > networking world. Routers and switches are very similar in > > > some ways, > > > > yet subtly different in others. In the mid-eighties, > > > arguments raged > > > > about which was "better". (Back then switches were > called bridges.) > > > > Today we understand that any large network > infrastructure needs to > > > > include both. > > > > > > > > Some people would quibble with my definitions. Some would > > > argue that > > > > if you run NFS over TCP/IP over fibre channel, then that is > > > a SAN. I > > > > figure that we didn't invent new names for NFS over 100bT, > > > FDDI, ATM, > > > > or Gigabit -- what's so special about fibre channel? > > > Others argue that > > > > a SAN is any network dedicated entirely to storage traffic. > > > Does this > > > > mean a SAN turns to a NAS if you fire up a telnet on > it? A year ago > > > > these definitions were less settled, but today I think > these are the > > > > definitions that are emerging. > > > > > > > > With these definitions, the key difference is between > file system > > > > protocols and raw disk protocols. With NAS, the filesystem > > > runs on the > > > > storage system itself. (For NetApp, that's WAFL.) > With SAN, the > > > > filesystem runs on the hosts attached to the storage system. > > > > > > > > People familiar with EMC will recognize this model. With an EMC > > > > Symetrix, you get a big box of RAID protected disks, and > > > you can attach > > > > multiple hosts to that one box. Thus, I view SAN as > "open EMC". On > > > > the one hand, SAN could be viewed as a threat to EMC. But > > > on the other > > > > hand, EMC has a solid history of products -- software and > > > hardware -- > > > > based on this storage model, which means that they ought to > > > have a real > > > > head-start on SAN. I think the SAN vendors have a great > > > opportunity if > > > > they really can create "open EMC". After all, EMC is doing > > > $4 Billion > > > > a year in business with this model! > > > > > > > > In any case, I think NAS has some serious advantages over > > > SAN. One big > > > > issue with running the file system on the separate > hosts is that it > > > > makes true data sharing very difficult. The on-disk byte > > > formats are > > > > very different between UNIX and NT, and even between > > > different flavors > > > > of UNIX. Given the differences between UNIT and NT filesystem > > > > semantics, it's no surprise that the on-disk format is > > > different. With > > > > NAS, the filesystem is by the data, so it can convert > the on-disk > > > > format into an industry standard protocol appropriate to > > > the host that > > > > wants the data: NFS for UNIX, NT Shares for NT, and HTTP for web > > > > browsers. > > > > > > > > It is technically possible to build a "global file system" > > > that has a > > > > common on-disk format for multiple operating systems, > thus allowing > > > > NAS-like sharing for SAN. However, I'm skeptical of that > > > model. First > > > > of all, there are no such products today. And even > when they become > > > > available, sharing won't be based on industry standards. A > > > system will > > > > only be able to participate in sharing if it runs a > special SAN file > > > > system from a particular vendor. My believe is that the > > > lesson of 20 > > > > years of networking is that we build heterogeneous > environments by > > > > creating open standards and allowing everyone to support > > > them -- not by > > > > saying that anyone who wants to participate in sharing > must buy code > > > > from one particular company. > > > > > > > > Having said all this, though, NetApp does have plans to > leverage SAN > > > > technology. We have announced an OEM agreement with > > > Brocade. Here are > > > > the benefits we expect to achieve: > > > > > > > > - Attach to lots of disks. > > > > > > > > A fibre channel loop is limited to 127 drives, > but with a 16 > > > > port fibre channel switch, the limit goes to almost 2000 > > > > drives. With additional switches, you can > support an arbitrary > > > > number of drives. > > > > > > > > - Share tape drives between multiple servers. > > > > > > > > I believe that backup applications are the > killer application > > > > for SAN. When people buy expensive tape > jukeboxes, they want > > > > to be able to attach them to multiple servers. > With today's > > > > SCSI tapes that's a painful job. But with > fibre channel and > > > > fibre channel switches, it's easy to connect a > single tape to > > > > any number of hosts. > > > > > > > > - Disaster Tolerant Solutions > > > > > > > > Fibre channel switches can do WAN tunneling to provide > > > > connections to remote disks without any changes > to the server > > > > code. This makes it much easier to build > remote mirrors for > > > > data replication and disaster protection. > > > > > > > > In conclusion, I believe that NAS and SAN are both important > > > > technologies, and the any large data infrastructure in the > > > future will > > > > include both. I believe that NAS focuses on issues > associated with > > > > users, applications, and data sharing. I believe that SAN > > > focuses on > > > > issues associated with disk and tape devices, > connection to large > > > > numbers of them, and moving data back and forth between > > > them. In the > > > > long run, any large environment will have both sets of > > > issues, and will > > > > need both types of solutions. In the short run, NAS makes > > > good sense > > > > as a starting point, because it is here today and the > standards it > > > > relies on have been solid for over a decade. > > > > > > > > I'll go ahead an post the whole paper when I finish, but I still > > > > thought it might be useful to share my thoughts in this brief, > > > > half-baked form. > > > > > > > > Dave > > > >

2 1

F740 cluster
by Jason D. Kelleher 06 Apr '99

06 Apr '99

Anybody out there have an F740 cluster w/ 18GB drives running in production? If so, I have a couple of questions I'm hoping you can answer: 1) Have any problems loosing routes over a virtual interface? If you physically disconect all the links in a trunk it drops all routes which use that interface, even static routes w/ routed off. 2) Have you been able to takeover/giveback a virtual interface? Treating it like a "normal" interface doesn't seem to work and I haven't been able to find any special instructions for this. 3) Any unexplained panics? We had one running 5.2.1 in takeover mode -- didn't even drop core. Another attributed to a bad disk while we were running 5.2.1P1. We're now at 5.2.1P2 after several ndmpd hangs (didn't bring down the filer), seems to be stable. 4) How many disk failures did you see within the first 6 months of getting the filer? (looks like we're on 2 of 24 in 3 months and we haven't even used them all yet). 5) Do you have the name and address of the person in charge of the hold muzac for the support line? The 5 minute loop of garbage is really ramping up my aggression level and I'd prefer to take it out on the person responsible. I have a couple of open calls with support I'm waiting for status on, but I'm not calling them because if I have to listen to that damned muzac tape loop around one more time I'm going to kill someone. Since my office-mate is closest it would probably be him... and that would just leave me more work to do. (Moving bodies, digging holes, and handling all his projects...) Any help would be greatly appreciated. jason --- Jason D. Kelleher kelleher(a)susq.com Susquehanna Partners, G.P. 610.617.2721 (voice) 401 City Line Ave, Suite 220 610.617.2916 (fax) Bala Cynwyd, PA 19004-1122

4 4

Re: Muzak (was F740 cluster)
by sirbruce＠ix.netcom.com 06 Apr '99

06 Apr '99

>Great! Does this mean we get to hear David Hitz's Greatest Hitz? I suddenly have this weird image of Dave Hitz doing Christian Slater's part in _Pump Up The Volume_. Bruce

1 0

Re: IRCACHE Web Cache Bakeoff?
by sirbruce＠ix.netcom.com 06 Apr '99

06 Apr '99

On 04/05/99 15:46:32 you wrote: >NetApp is pleased with its performance in the NLANR bake-off and are >sharing our results with customers and potential customers under NDA. >We have decided not to publicly release our results at this time >because of the late withdrawal from the bake-off by the other major >caching vendors (Cisco and Inktomi). Network Appliance does not want >to disclose detailed performance numbers to our competitors without >having access to their results. Based on past experience we believe >that competitors are likely to compare apples to oranges and >selectively quote our results as it suits their requirements. If this is true, it's completely nonsensical. Other vendors *did* have their results published. Why should you care only about the couple that didn't? Furthermore, the same logic means you shouldn't be releasing any NFS or CIFS performance numbers either, since not all of your competitors do. (EMC, for example, still hasn't submitted new SFS97 results and continues to point to older benchmarks.) Yet this hasn't stopped you in the past. Is this a new policy, or one which you only want to apply to Polygraph? Bruce

1 0

snapshot doc
by dnb＠ccs.neu.edu 05 Apr '99

05 Apr '99

Howdy- Just so I feel like I've actually added some signal to today's discussion, I wanted to point folks at: http://www.ccs.neu.edu/groups/systems/howto/howto-snapshot.html This is a document I wrote when we first got our NetApp explaining snapshots to our user community and how they are used for online file recovery. Feel free to use it with local modifications if you ever need to publicize same. Respectfully, David N. Blank-Edelman Director of Technology College of Computer Science Northeastern University

1 0

Re: IRCACHE Web Cache Bakeoff?
by David Drum 05 Apr '99

05 Apr '99

Hello everyone, I have received a completely satisfactory answer from my sales rep. about why NetApp pulled out of the bakeoff. Hopefully someone from NetApp will respond and quench the rumor mill. It would be inappropriate of me to forward the email I received. Regards, David david(a)mu.org

1 0

IRCACHE Web Cache Bakeoff?
by sirbruce＠ix.netcom.com 05 Apr '99

05 Apr '99

Can anyone at Netapp tell us what happened with the ircache web cache bakeoff? Netapp attended and then declined to have their results published. Many other companies backed out as well, such as Cacheflow and IBM. Rumor has it that netcache was outperformed; was the benchmark unrealistic or did Netapp withdraw for other reasons? Bruce

2 1

Re: F740 cluster
by dnb＠ccs.neu.edu 05 Apr '99

05 Apr '99

"Jason D. Kelleher" <kelleher(a)susq.com> writes: > 5) Do you have the name and address of the person in charge of > the hold muzac for the support line? The 5 minute loop of > garbage is really ramping up my aggression level and I'd > prefer to take it out on the person responsible. Hear, hear! I've said almost the exact same words at every support person I've ever spoken to (and some people higher). I even offered to purchase the music myself. In the end, I decided against hurting someone because I decided it would be too embarrassing to have to describe the situation to others in federal prison: "So what are you in for?" "Murder." "And you?" "Armed robbery." "What about you?" "Kenny G. It all started with this NetApp support call..." It's not like they wouldn't understand, but still. Respectfully, David N. Blank-Edelman Director of Technology College of Computer Science Northeastern University

2 1

Muzak (was F740 cluster)
by Skinner, Alistair 05 Apr '99

05 Apr '99

Hi Jason, My name is Alistair Skinner. I run the global support center at Network Appliance, and was passed your comments concerning the "hold muzac". Thanks for the feedback, I would guess there are two issues: (1) why do we have to keep you on hold so long? We have put in place several initiatives around improving our staffing, scheduling, training, processes, and systems which we expect to improve service levels over time. In the last few months we have hired 10 more staff, defined a more rigorous initial training program, and worked to eliminate our backlog of cases. Although our average holdtime is now 3 minutes, we know there are quite a few instances where customer have had to wait much longer, so we will keep tuning our goals, scheduling, and processes to provide more consistent response. (2) if we do have to keep you on hold, can't we play something better? We knew people were having trouble with Kenny G, so we changed recently to a longer sample of different music. However, it sounds like that isn't cutting it either, and so we are purchasing an audio feed card for our Aspect phone system. This will allow us to have a live feed instead of a loop. Based on your feedback, I'll see what we can do about accelerating its delivery and installation. Alistair Alistair Skinner Network Appliance alistair.skinner(a)netapp.com 408-367-3869 -----Original Message----- From: Jason D. Kelleher [mailto:kelleher@susq.com] Sent: Monday, April 05, 1999 10:50 AM 5) Do you have the name and address of the person in charge of the hold muzac for the support line? The 5 minute loop of garbage is really ramping up my aggression level and I'd prefer to take it out on the person responsible.

1 0

RE: NAC SAN
by Adams, Christian 05 Apr '99

05 Apr '99

Hey Ray Chen - Thanks for the clarification of Dave Hitz's position. /Christian Adams EMC > -----Original Message----- > From: Chen, Ray [SMTP:ray.chen@netapp.com] > Sent: Sunday, April 04, 1999 10:19 AM > To: 'Adams, Christian'; Hitz, Dave; willh(a)infi.net > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > Subject: RE: NAC SAN > > Um, because the distinctive characteristic of SAN is that > clients can write directly to the storage devices bypassing > any server? This is possible only if a storage-level (e.g. > block-level) protocol like SCSI runs on the wire. (Actually, > I'm lying here. There are two ways of enabling this. One is > to use a really simple protocol like SCSI. The other is to > implement an industry-standard clustered filesystem with the > disk drives/storage devices as an intelligent component of > said filesystem. A very complex protocol whose viability > and practicality is an issue in its own right. The rest of this > mail assumes a block-level SAN protocol.) > > As folks have pointed out, SAN has its advantages and disadvantages. > It eliminates the backplane of server (unless you have a RAID controller > in front of the drives :-) from the I/O path, eliminating a possible > I/O bottleneck. But there is no server to protect the integrity > of the data and even if there is, the protocol is so low-level in > a SAN, the server has no effective way of detecting when a client > has gone insane (maliciously or not) and is scribbling gibberish > on the drives. So trust becomes a definite issue. > > Personally, I see data sharing using SANS as making sense > in small workgroup environments and what I call "machine room > clusters" -- a SAN in a machine room supporting a dedicated > cluster. Typically you'd do the latter because scaling via > a cluster is cheaper (and more available) than scaling via MP. > But both are situations where you are willing to trust all the > machines in the cluster because they are all firmly controlled by > the group whose storage/data is being shared. > > Ray Chen > rcc(a)netapp.com > > > -----Original Message----- > > From: Adams, Christian [mailto:Adams_Christian@emc.com] > > Sent: Tuesday, March 30, 1999 10:14 AM > > To: 'Dave Hitz'; willh(a)infi.net > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > Subject: RE: NAC SAN > > > > > > Hey Dave Hitz - > > > > Why do you focus on existing protocols in your discussion? > > This indicates > > that the introduction of Fibrechanel with either encapsulated SCSI or > > Ethernet is the critical event in the creation of SAN or NAS. > > Further in > > your discussion, you draw a parallel to switches and routers > > which indicate > > that you may be thinking in terms of shared versus segmented > > data flows. > > And yet in all of your discussion, you do not address the issues of > > management, integration, or value. > > > > Loved your EMC bit, > > /Christian Adams > > EMC > > > > > -----Original Message----- > > > From: Dave Hitz [SMTP:hitz@netapp.com] > > > Sent: Saturday, March 27, 1999 1:41 PM > > > To: willh(a)infi.net > > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > > Subject: Re: NAC SAN > > > > > > > Tom's statement below (to a [NetApp] SAN when we build > > one) brings up a > > > > familiar argument between myself and my boss. My boss > > contends that > > > > NetApp is NOT capable of migrating to a "real SAN." > > > > > > I've been delinquent in my duty of writing a NAS and SAN paper. I'm > > > about half way done. Whem I'm done, I'll post it here, probably in > > > early form, so that you guys can critique it for me. > > > > > > Very briefly, I believe that NAS and SAN are different but related > > > technologies, and that the definitions are as follows: > > > > > > NAS (Network Attached Storage) > > > Storage accessed over TCP/IP, using industry standard file > > > sharing protocols like NFS, HTTP and Windows Networking. > > > > > > SAN (Storage Area Network) > > > Storage accessed over a Fibre Channel switching fabric, > > > using encapsulated SCSI. > > > > > > I also think that both NAS and SAN are useful, and that -- > > in the long > > > run -- both will be critical components in any large data > > > infrastructure. I think there's a strong analogy to the TCP/IP > > > networking world. Routers and switches are very similar in > > some ways, > > > yet subtly different in others. In the mid-eighties, > > arguments raged > > > about which was "better". (Back then switches were called bridges.) > > > Today we understand that any large network infrastructure needs to > > > include both. > > > > > > Some people would quibble with my definitions. Some would > > argue that > > > if you run NFS over TCP/IP over fibre channel, then that is > > a SAN. I > > > figure that we didn't invent new names for NFS over 100bT, > > FDDI, ATM, > > > or Gigabit -- what's so special about fibre channel? > > Others argue that > > > a SAN is any network dedicated entirely to storage traffic. > > Does this > > > mean a SAN turns to a NAS if you fire up a telnet on it? A year ago > > > these definitions were less settled, but today I think these are the > > > definitions that are emerging. > > > > > > With these definitions, the key difference is between file system > > > protocols and raw disk protocols. With NAS, the filesystem > > runs on the > > > storage system itself. (For NetApp, that's WAFL.) With SAN, the > > > filesystem runs on the hosts attached to the storage system. > > > > > > People familiar with EMC will recognize this model. With an EMC > > > Symetrix, you get a big box of RAID protected disks, and > > you can attach > > > multiple hosts to that one box. Thus, I view SAN as "open EMC". On > > > the one hand, SAN could be viewed as a threat to EMC. But > > on the other > > > hand, EMC has a solid history of products -- software and > > hardware -- > > > based on this storage model, which means that they ought to > > have a real > > > head-start on SAN. I think the SAN vendors have a great > > opportunity if > > > they really can create "open EMC". After all, EMC is doing > > $4 Billion > > > a year in business with this model! > > > > > > In any case, I think NAS has some serious advantages over > > SAN. One big > > > issue with running the file system on the separate hosts is that it > > > makes true data sharing very difficult. The on-disk byte > > formats are > > > very different between UNIX and NT, and even between > > different flavors > > > of UNIX. Given the differences between UNIT and NT filesystem > > > semantics, it's no surprise that the on-disk format is > > different. With > > > NAS, the filesystem is by the data, so it can convert the on-disk > > > format into an industry standard protocol appropriate to > > the host that > > > wants the data: NFS for UNIX, NT Shares for NT, and HTTP for web > > > browsers. > > > > > > It is technically possible to build a "global file system" > > that has a > > > common on-disk format for multiple operating systems, thus allowing > > > NAS-like sharing for SAN. However, I'm skeptical of that > > model. First > > > of all, there are no such products today. And even when they become > > > available, sharing won't be based on industry standards. A > > system will > > > only be able to participate in sharing if it runs a special SAN file > > > system from a particular vendor. My believe is that the > > lesson of 20 > > > years of networking is that we build heterogeneous environments by > > > creating open standards and allowing everyone to support > > them -- not by > > > saying that anyone who wants to participate in sharing must buy code > > > from one particular company. > > > > > > Having said all this, though, NetApp does have plans to leverage SAN > > > technology. We have announced an OEM agreement with > > Brocade. Here are > > > the benefits we expect to achieve: > > > > > > - Attach to lots of disks. > > > > > > A fibre channel loop is limited to 127 drives, but with a 16 > > > port fibre channel switch, the limit goes to almost 2000 > > > drives. With additional switches, you can support an arbitrary > > > number of drives. > > > > > > - Share tape drives between multiple servers. > > > > > > I believe that backup applications are the killer application > > > for SAN. When people buy expensive tape jukeboxes, they want > > > to be able to attach them to multiple servers. With today's > > > SCSI tapes that's a painful job. But with fibre channel and > > > fibre channel switches, it's easy to connect a single tape to > > > any number of hosts. > > > > > > - Disaster Tolerant Solutions > > > > > > Fibre channel switches can do WAN tunneling to provide > > > connections to remote disks without any changes to the server > > > code. This makes it much easier to build remote mirrors for > > > data replication and disaster protection. > > > > > > In conclusion, I believe that NAS and SAN are both important > > > technologies, and the any large data infrastructure in the > > future will > > > include both. I believe that NAS focuses on issues associated with > > > users, applications, and data sharing. I believe that SAN > > focuses on > > > issues associated with disk and tape devices, connection to large > > > numbers of them, and moving data back and forth between > > them. In the > > > long run, any large environment will have both sets of > > issues, and will > > > need both types of solutions. In the short run, NAS makes > > good sense > > > as a starting point, because it is here today and the standards it > > > relies on have been solid for over a decade. > > > > > > I'll go ahead an post the whole paper when I finish, but I still > > > thought it might be useful to share my thoughts in this brief, > > > half-baked form. > > > > > > Dave > >

1 0

RE: F740 cluster
by Sam Schorr 05 Apr '99

05 Apr '99

Jason, We do have 2 F740's clustered, the first one (our "original" filer) has 2 racks of 9 gig drives, while the newer cluster partner has 2 racks of 18gig drives. We do not use virtual interfaces, so I'm not sure what relevance our experience is. We did have quite an adventure, with many "downs", all on the first filer, although none turned out to be cluster related. We have not had any drive failures to date. We are on 5.2.1P2D6, primarily because we are an exclusive CIFS shop, and due to some networking problems, the filer would reboot itself (an problem with WAFL handling CIFS that sends commands out of sequence). Netapp worked hard and vigorously to retrofit a patch that is in 5.3.1 (and thus not released) into 5.2.1 (thus the D6). We did have some initial issue with the disk shelves on our 1st filer that required swapping them to get clustering to work. All clustering commands have worked as advertised, but as we only have one physical interface running at this time, I'm not sure it helps you. --srs -----Original Message----- From: Jason D. Kelleher [mailto:kelleher@susq.com] Sent: Monday, April 05, 1999 10:50 AM To: toasters(a)mathworks.com Subject: F740 cluster Anybody out there have an F740 cluster w/ 18GB drives running in production? If so, I have a couple of questions I'm hoping you can answer: 1) Have any problems loosing routes over a virtual interface? If you physically disconect all the links in a trunk it drops all routes which use that interface, even static routes w/ routed off. 2) Have you been able to takeover/giveback a virtual interface? Treating it like a "normal" interface doesn't seem to work and I haven't been able to find any special instructions for this. 3) Any unexplained panics? We had one running 5.2.1 in takeover mode -- didn't even drop core. Another attributed to a bad disk while we were running 5.2.1P1. We're now at 5.2.1P2 after several ndmpd hangs (didn't bring down the filer), seems to be stable. 4) How many disk failures did you see within the first 6 months of getting the filer? (looks like we're on 2 of 24 in 3 months and we haven't even used them all yet). 5) Do you have the name and address of the person in charge of the hold muzac for the support line? The 5 minute loop of garbage is really ramping up my aggression level and I'd prefer to take it out on the person responsible. I have a couple of open calls with support I'm waiting for status on, but I'm not calling them because if I have to listen to that damned muzac tape loop around one more time I'm going to kill someone. Since my office-mate is closest it would probably be him... and that would just leave me more work to do. (Moving bodies, digging holes, and handling all his projects...) Any help would be greatly appreciated. jason --- Jason D. Kelleher kelleher(a)susq.com Susquehanna Partners, G.P. 610.617.2721 (voice) 401 City Line Ave, Suite 220 610.617.2916 (fax) Bala Cynwyd, PA 19004-1122

1 0

Re: F740 cluster
by Jason D. Kelleher 05 Apr '99

05 Apr '99

In message <199904051837.LAA08270(a)sas-pc.netapp.com>, Scott Schoenthal write s: >> 2) Have you been able to takeover/giveback a virtual interface? >> Treating it like a "normal" interface doesn't seem to work >> and I haven't been able to find any special instructions >> for this. > >Virtual interface (vif, Etherchannel, link aggregation, your-name-here) >takeover/giveback is not supported in 5.2.x > >The 5.3 release supports virtual interface takeover/giveback. > >/sas Thanks, I'll write that little gem in my 5.2.1 manual... Did anyone consider putting this info in the System Administrator's Guide? the Cluster Conversion Guide, how 'bout somewhere on the NOW site? I try to be a good little user and RTFM, but if the FM doesn't have the info... :( jason

1 0

RE: IRCACHE Web Cache Bakeoff?
by sirbruce＠ix.netcom.com 05 Apr '99

05 Apr '99

On 04/05/99 20:07:01 you wrote: > > >Actually Netchache was doing very well but since no competition >was present the performance figures will not show if A or B >was the best. My own opinion is that Netapp will present the figures >but first do a thorough check to see if the competition will >have an opinon why they didn't enter and if they will kill the outcome. >Hence why didn't Inktomi, Cacheflow, etc enter? But some competition *was* present. The report was already released. http://bakeoff.ircache.net/ Bruce

1 0

RE: F740 cluster
by Smith, Gil 05 Apr '99

05 Apr '99

Hi Jason, I am just going to comment on a few of your items below: > 2) Have you been able to takeover/giveback a virtual > interface? > Treating it like a "normal" interface doesn't seem to work > and I haven't been able to find any special instructions > for this. I believe 5.3 includes support for this ala "pvif" which is policy based virtual interface (superset of vif). 5.3 should be out any day now and the details of the "pvif" will be documented in the usual places (*.pdf files available in the software library section of now.netapp.com for current holders of the software subscription service option). > I have a couple of open calls with support I'm waiting for status > on, but I'm not calling them because if I have to listen to that > damned muzac tape loop around one more time I'm going to kill > someone. Since my office-mate is closest it would probably be > him... and that would just leave me more work to do. (Moving > bodies, digging holes, and handling all his projects...) > > Any help would be greatly appreciated. I am truly sorry you have had to sit and listen to this music. It is certainly not the intention of anyone at Network Appliance that this happens. I have sent you some names via a separate email but will comment in general to this list. If you *ever* have a support gripe, concern, or feedback please call and ask to speak with the technical support shift manager. We value your business. Regards, Gil

1 0

RE: IRCACHE Web Cache Bakeoff?
by Jonas Jonsson 05 Apr '99

05 Apr '99

Actually Netchache was doing very well but since no competition was present the performance figures will not show if A or B was the best. My own opinion is that Netapp will present the figures but first do a thorough check to see if the competition will have an opinon why they didn't enter and if they will kill the outcome. Hence why didn't Inktomi, Cacheflow, etc enter? /// Best regards, Jonas -----Original Message----- From: sirbruce(a)ix.netcom.com [mailto:sirbruce@ix.netcom.com] Sent: den 5 april 1999 06:14 To: toasters(a)mathworks.com Subject: IRCACHE Web Cache Bakeoff? Can anyone at Netapp tell us what happened with the ircache web cache bakeoff? Netapp attended and then declined to have their results published. Many other companies backed out as well, such as Cacheflow and IBM. Rumor has it that netcache was outperformed; was the benchmark unrealistic or did Netapp withdraw for other reasons? Bruce

1 0

RE: IRCACHE Web Cache Bakeoff?
by Mohler, Jeff 05 Apr '99

05 Apr '99

Surely it feels like a failure, but we'll see. The 'spin' on the situation is not good. > -----Original Message----- > From: sirbruce(a)ix.netcom.com [SMTP:sirbruce@ix.netcom.com] > Sent: Sunday, April 04, 1999 11:14 PM > To: toasters(a)mathworks.com > Subject: IRCACHE Web Cache Bakeoff? > > Can anyone at Netapp tell us what happened with the > ircache web cache bakeoff? Netapp attended and then > declined to have their results published. Many other > companies backed out as well, such as Cacheflow and IBM. > Rumor has it that netcache was outperformed; was the > benchmark unrealistic or did Netapp withdraw for other > reasons? > > Bruce

1 0

RE: NAC SAN
by Chen, Ray 05 Apr '99

05 Apr '99

Um, because the distinctive characteristic of SAN is that clients can write directly to the storage devices bypassing any server? This is possible only if a storage-level (e.g. block-level) protocol like SCSI runs on the wire. (Actually, I'm lying here. There are two ways of enabling this. One is to use a really simple protocol like SCSI. The other is to implement an industry-standard clustered filesystem with the disk drives/storage devices as an intelligent component of said filesystem. A very complex protocol whose viability and practicality is an issue in its own right. The rest of this mail assumes a block-level SAN protocol.) As folks have pointed out, SAN has its advantages and disadvantages. It eliminates the backplane of server (unless you have a RAID controller in front of the drives :-) from the I/O path, eliminating a possible I/O bottleneck. But there is no server to protect the integrity of the data and even if there is, the protocol is so low-level in a SAN, the server has no effective way of detecting when a client has gone insane (maliciously or not) and is scribbling gibberish on the drives. So trust becomes a definite issue. Personally, I see data sharing using SANS as making sense in small workgroup environments and what I call "machine room clusters" -- a SAN in a machine room supporting a dedicated cluster. Typically you'd do the latter because scaling via a cluster is cheaper (and more available) than scaling via MP. But both are situations where you are willing to trust all the machines in the cluster because they are all firmly controlled by the group whose storage/data is being shared. Ray Chen rcc(a)netapp.com > -----Original Message----- > From: Adams, Christian [mailto:Adams_Christian@emc.com] > Sent: Tuesday, March 30, 1999 10:14 AM > To: 'Dave Hitz'; willh(a)infi.net > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > Subject: RE: NAC SAN > > > Hey Dave Hitz - > > Why do you focus on existing protocols in your discussion? > This indicates > that the introduction of Fibrechanel with either encapsulated SCSI or > Ethernet is the critical event in the creation of SAN or NAS. > Further in > your discussion, you draw a parallel to switches and routers > which indicate > that you may be thinking in terms of shared versus segmented > data flows. > And yet in all of your discussion, you do not address the issues of > management, integration, or value. > > Loved your EMC bit, > /Christian Adams > EMC > > > -----Original Message----- > > From: Dave Hitz [SMTP:hitz@netapp.com] > > Sent: Saturday, March 27, 1999 1:41 PM > > To: willh(a)infi.net > > Cc: toasters(a)mathworks.com; barsellt(a)infi.net > > Subject: Re: NAC SAN > > > > > Tom's statement below (to a [NetApp] SAN when we build > one) brings up a > > > familiar argument between myself and my boss. My boss > contends that > > > NetApp is NOT capable of migrating to a "real SAN." > > > > I've been delinquent in my duty of writing a NAS and SAN paper. I'm > > about half way done. Whem I'm done, I'll post it here, probably in > > early form, so that you guys can critique it for me. > > > > Very briefly, I believe that NAS and SAN are different but related > > technologies, and that the definitions are as follows: > > > > NAS (Network Attached Storage) > > Storage accessed over TCP/IP, using industry standard file > > sharing protocols like NFS, HTTP and Windows Networking. > > > > SAN (Storage Area Network) > > Storage accessed over a Fibre Channel switching fabric, > > using encapsulated SCSI. > > > > I also think that both NAS and SAN are useful, and that -- > in the long > > run -- both will be critical components in any large data > > infrastructure. I think there's a strong analogy to the TCP/IP > > networking world. Routers and switches are very similar in > some ways, > > yet subtly different in others. In the mid-eighties, > arguments raged > > about which was "better". (Back then switches were called bridges.) > > Today we understand that any large network infrastructure needs to > > include both. > > > > Some people would quibble with my definitions. Some would > argue that > > if you run NFS over TCP/IP over fibre channel, then that is > a SAN. I > > figure that we didn't invent new names for NFS over 100bT, > FDDI, ATM, > > or Gigabit -- what's so special about fibre channel? > Others argue that > > a SAN is any network dedicated entirely to storage traffic. > Does this > > mean a SAN turns to a NAS if you fire up a telnet on it? A year ago > > these definitions were less settled, but today I think these are the > > definitions that are emerging. > > > > With these definitions, the key difference is between file system > > protocols and raw disk protocols. With NAS, the filesystem > runs on the > > storage system itself. (For NetApp, that's WAFL.) With SAN, the > > filesystem runs on the hosts attached to the storage system. > > > > People familiar with EMC will recognize this model. With an EMC > > Symetrix, you get a big box of RAID protected disks, and > you can attach > > multiple hosts to that one box. Thus, I view SAN as "open EMC". On > > the one hand, SAN could be viewed as a threat to EMC. But > on the other > > hand, EMC has a solid history of products -- software and > hardware -- > > based on this storage model, which means that they ought to > have a real > > head-start on SAN. I think the SAN vendors have a great > opportunity if > > they really can create "open EMC". After all, EMC is doing > $4 Billion > > a year in business with this model! > > > > In any case, I think NAS has some serious advantages over > SAN. One big > > issue with running the file system on the separate hosts is that it > > makes true data sharing very difficult. The on-disk byte > formats are > > very different between UNIX and NT, and even between > different flavors > > of UNIX. Given the differences between UNIT and NT filesystem > > semantics, it's no surprise that the on-disk format is > different. With > > NAS, the filesystem is by the data, so it can convert the on-disk > > format into an industry standard protocol appropriate to > the host that > > wants the data: NFS for UNIX, NT Shares for NT, and HTTP for web > > browsers. > > > > It is technically possible to build a "global file system" > that has a > > common on-disk format for multiple operating systems, thus allowing > > NAS-like sharing for SAN. However, I'm skeptical of that > model. First > > of all, there are no such products today. And even when they become > > available, sharing won't be based on industry standards. A > system will > > only be able to participate in sharing if it runs a special SAN file > > system from a particular vendor. My believe is that the > lesson of 20 > > years of networking is that we build heterogeneous environments by > > creating open standards and allowing everyone to support > them -- not by > > saying that anyone who wants to participate in sharing must buy code > > from one particular company. > > > > Having said all this, though, NetApp does have plans to leverage SAN > > technology. We have announced an OEM agreement with > Brocade. Here are > > the benefits we expect to achieve: > > > > - Attach to lots of disks. > > > > A fibre channel loop is limited to 127 drives, but with a 16 > > port fibre channel switch, the limit goes to almost 2000 > > drives. With additional switches, you can support an arbitrary > > number of drives. > > > > - Share tape drives between multiple servers. > > > > I believe that backup applications are the killer application > > for SAN. When people buy expensive tape jukeboxes, they want > > to be able to attach them to multiple servers. With today's > > SCSI tapes that's a painful job. But with fibre channel and > > fibre channel switches, it's easy to connect a single tape to > > any number of hosts. > > > > - Disaster Tolerant Solutions > > > > Fibre channel switches can do WAN tunneling to provide > > connections to remote disks without any changes to the server > > code. This makes it much easier to build remote mirrors for > > data replication and disaster protection. > > > > In conclusion, I believe that NAS and SAN are both important > > technologies, and the any large data infrastructure in the > future will > > include both. I believe that NAS focuses on issues associated with > > users, applications, and data sharing. I believe that SAN > focuses on > > issues associated with disk and tape devices, connection to large > > numbers of them, and moving data back and forth between > them. In the > > long run, any large environment will have both sets of > issues, and will > > need both types of solutions. In the short run, NAS makes > good sense > > as a starting point, because it is here today and the standards it > > relies on have been solid for over a decade. > > > > I'll go ahead an post the whole paper when I finish, but I still > > thought it might be useful to share my thoughts in this brief, > > half-baked form. > > > > Dave >

2 1

NFS invitation
by tkaczma＠gryf.net 05 Apr '99

05 Apr '99

I believe someone invited us to a NFS v4 discussion sometime back. However, somehow the message hit the bit bucket on my machine. Could someone forward me a copy of that letter or direct me to a site mentioned inside? I would really like to see ACLs in NFS v4. I think this is one of few advantages that CIFS, AFS, and others have over NFS. Tom

2 1

RE: Slow sequential disk reads on F740
by Brian Meek 04 Apr '99

04 Apr '99

Setting options nfs.udp.xfersize 32768 helped a lot, now were seeing ~25 MB/sec transfers to and from the filer. Why isn't this the default? Brian -----Original Message----- From: Smith, Gil [mailto:gil.smith@netapp.com] Sent: Friday, April 02, 1999 1:38 PM To: toasters(a)mathworks.com Subject: FW: Slow sequential disk reads on F740 -----Original Message----- From: Smith, Gil Sent: Friday, April 02, 1999 4:18 PM To: 'toaster(a)mathworks.com' Subject: RE: Slow sequential disk reads on F740 Hi, You might want to try, filer> options nfs.udp.xfersize 32768 which is a hidden option (doesn't show up when you type options with no arguments). This might be especially usefull in Gigabit Ethernet environments. Regards, Gil -----Original Message----- From: Brian Tao [mailto:taob@risc.org] Sent: Thursday, April 01, 1999 5:02 PM To: Brian Meek Cc: toasters(a)mathworks.com Subject: RE: Slow sequential disk reads on F740 On Thu, 1 Apr 1999, Brian Meek wrote: > > I'm not sure this is related, but if you do a nfsstat -m on the client, does > it show the rsize to be 8192 or 32768? Interesting... I was forcing rsize and wsize to 32768 on the mount command line just to be pedantic, but this is what I'm seeing (output edited for brevity): # mount -o vers=3,proto=udp,rsize=32768,wsize=32768 e2.j:/ /j # mount /j on e2.j:/ vers=3/proto=udp/rsize=32768/wsize=32768/remote on Thu Apr 1 16:51:27 1999 # nfsstat -m /j from e2.j:/ Flags: vers=3,proto=udp,sec=sys,hard,intr,link,symlink,acl,rsize=8192,wsize=8192,re tran s=5 > When reading data from our NetApps using our E450's we don't see > anywhere near the same performance as when we read data from our > E450's with our E450's using NFS3 over udp. I've always attributed > this to the client code forcing an rsize of 8192 when connecting to > the filler, and allowing a rsize of 32768 when connecting to > anything else (Sun, Dec, SGI). Has anyone else seen this? It sounds like the NFS server dictates the parameters. I can connect with 32K rsize/wsize using UDP transport between Solaris servers, but only 8K between a Sun and an Netapp, and 16K between a Sun and a FreeBSD machine acting as the server. TCP transport allows me to use 32K block sizes in any server. -- Brian Tao (BT300, taob(a)risc.org) "Though this be madness, yet there is method in't"

3 2

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters