Technically Speaking

The last couple days (minus travel) I have been in Fort Collins CO learning about Converged Infrastructure, 3PAR, VirtualSystem and other things HP is doing in storage.  I got the invite about a month ago and having attended other events in the past, I was all in.

I am not sure if the knowledge, and technical learning or the people and networking I get the opportunity to do is the best part.  I will admit, the technical things I liked the most were the ones I knew the least about going in.  My organization is small, and we have bigger storage than we did even two years ago, but some of the storage I have been learning about lately is definitely in the Big Enterprise category.

3PAR

I knew 3PAR was a service provider focussed storage company that HP acquired.  That is about it.  I think I also knew they used yellow bezels.  Some of the programming and data handling software that 3PAR is using to efficiently handle data is amazing.  A good amount of the decisions made (once the user decides where the data should be placed) are handled on the backend in the array.

Once the offload happens, 3PAR has a set of algorithms and software called ASIC that get to work on the commands passed in by an admin.  The compute work is also handled in storage and the whole thing is very fast.  The hands on lab was a great experience, allowing me to play with a 3PAR array, something I will likely not get to do in the real world.

It would be cool if HP could make some of these labs available via the web to allow further learning by us geeks who just have a desire to know as much as we can hold about storage.

I know that I havent covered everything we saw from 3PAR, but for me the ASIC discussion and the efficiency of the system were definitely the highlights.

Virtual System

I have always wanted a kick ass pile of compute in a box.  You know, the kind that falls from the sky and plugs in, bringing you an enterprise quality system.  HP is doing this on medium, large, and huge scale.  They sell certain levels of virtual system which brings compute, management compute, networking, and storage into one or more interconnected rack enclosures.  The customer works with HP and chooses the options they want, and the racks appear  (plus shipping of course). Even though the systems can be high to start, it is a ground up, quick way to take the mess out of a datacenter and get a clean, fast, product that just works.

Networking

The networking opportunity that came from this event, with both other bloggers (some of whom I had met before and some new faces) and HP employees was incredible.  I know that I will be reaching out to a few of the HP-ers to learn more about some of these products and to the other bloggers to help my network and general knowledge of storage technology grow.  I am amazed how much I have learned over the past few years both from attending events and meeting people.

Next up.

Once I get back to Wisconsin (boarding the plane in just under 2 hours) I will be knee deep in Tech On Tap.  That kicks off tomorrow and really is a similar event in style to what I have been doing this week.  On a smaller scale, since we selected a technology and found some smart people to come in and help us talk about it. As much as I am tired at the moment, I cant wait to see what tomorrow brings.

I know that I will also be chasing down a good digital recorder before I goto another event like this.  It is really hard to keep up sometimes and maybe getting notes that way and sculpting them into useful stuff after that is the way to do it.  I missed a good chunk of a virtualization integration session while fielding a couple of support calls, nobodys fault, but a good opportunity to record and review later.  The portions of the session I did catch were interesting to say the least.  I have a lot to learn about virtualization.

Thank you

I want to thank HP (@HPStorageguy and @beccataylor) for putting this event together and all the other HP-ers and bloggers who participated and helped me learn something new.

 So this week is storming right by and is capped off with a trip to Fort Collins CO to learn about 3Par and Lefthand and how HP does storage.  I am interested to learn more about both of these products, and even storage in general. There will be lots of useful discussion for sure which is the best part.  

Bringing bloggers together to discuss and learn about HP products and the tech that makes them work is great and I’m glad to participate.  Like others attending I have read some of the white papers covering both 3par and Lefthand but have no experience with them. No worries, I’m sure it is coming soon. After going to CO, next up is Tech on Tap I am very much looking forward to the first event and hope it is the first of many.

Time to pack and charge my iDevices.

Recently I have been looking for tools to better help manage the many things that are going on in my organization.  Mainly storage and devices attached to my network to help me keep tabs on just what is going on.  During a recent trip to Austin Tx for TechFieldDay I got another opportunity to hear from SolarWinds about some of the products they have been working on.

Storage Manager

This application is designed to provide insight into things going on with monitored servers and storage in a given environment.  When I got back from Austin, I was hoping to spend some time with this (and other) SolarWinds applications, and recently started down that path.  Because I just got things going, I cannot give a full and in depth review just yet, but will do so in the future.

The idea behind storage manager is a single pane of glass view into storage arrays and backup instances within your environment.  Initially I expected to see something that looks just at a select list of storage arrays and provides data via SNMP or access to the array.  Storage Manager does this, but it also allows me to configure agents on my Windows servers and monitor disk space and memory usage for each.

In addition to the overall storage monitoring, backups can also be watched and monitored from here.  This is something I discovered somewhat haphazardly in the initial tinkering phase.  It was quite a pleasant surprise, but will need further investigation.

In the initial demo of the product, I was rather blown away by all of the knobs available to look at an environment.  Now that I have gotten the ball rolling on testing and using SolarWinds in my environment, there is much to learn and lots to try.

The session in Austin where SolarWinds provided demos and information was very interesting and there was lots to see.  Of course getting it moving in my own environment will take some time and patience, but I’ll get there soon. Hopefully in time for next year’s budget cycle.

First impressions

The first thing I noticed when I started putting together the pieces of the SolarWinds puzzle is how well they work together.  Because of scheduling and issues with some of my test VMs I previously had trouble getting some other products from SolarWinds running correctly.  This time around, I started from scratch and all of the applications I am going to be testing installed easily and seem to work without errors.  Keep in mind, many of the tools and things are as yet untouched, but as I get more time to dig into them I am excited to see where they take me.

What am I hoping for?

Time to learn new ways to get swamped by data I think is first on my list.  Then a good opportunity to work with both the application(s) and the people behind them to really get good at understanding what they are exposing and how I can use that information to make decisions about technology going forward.

I realize I am not yet reviewing the application, and that’s OK.  It has been a long time coming and I am looking forward to getting my hands dirty soon.

Based on what I have seen so far I cannot wait to dig in further.

Fabio Rapposelli has a great post about SolarWinds and their presentation at TechFieldDay

Disclaimer: SolarWinds was a presenting sponsor at TechFieldDay, while they did host part of the event and help make it possible for me to attend, the reviews and posts here are my own.  

Because backing up data is something that goes without question. Sure care should be taken to select the right software and media to ensure your (or your organization’s) data is regularly backed up.  Imagine if there was a single solution allowing easy management and usage.

Symantec Backup Exec Appliance

Symantec Backup Exec is coming to a new single U / single SKU appliance that will manage your backups and restores.

The Symantec Backup Exec v3600 is designed to take the management pain out of backing up data.  By including all of the agent licenses for Windows and Exchange and SQL (among others) there is no guessing and ordering to ensure that the correct licensing exists.  The goal of the device is to simply do backups and restores after a little configuration.

Configuration

Getting the appliance up and running requires knowledge of your network and an available IP address for the device.  It does need to be on the same subnet as the other devices (the remote agents) that you will be backing up.  Once this is configured, build your backup jobs and get them scheduled.

Storage

This appliance will manage 5.5 TB of backup deduplicated data, which for many SMB organizations is plenty of storage space. Because the data is deduped the storage space is increased even further. The disk based nature of the appliance makes for improved backup and recovery speed. But be aware that in the first implementation there is no tape support. Backup data can be migrated to a USB drive or replicated to other Symantec Backup Exec appliances, the cloud through a few different hosts, or a USB drive.

The backup data will be deduplicated before it heads off to the cloud, based on the cost of bandwidth to get data onto a cloud service, this is kind of a big deal.  For an extremely small organization the cloud might be a good off-site option, but for many I feel that even deduped data will be to expensive to send there.  The inclusion of a cloud option may be a great decision going forward when cloud services get cheap enough to allow this type of thing.

Simple and all inclusive backup and easy restores appears to be the goal, but there are some caveats there as well.

Also, you cannot use SAN storage for backups from the device.  Once the internal storage is full, that’s all she wrote.  Adding the ability to use existing SAN storage once the internal storage gets to a certain percentage (say 80%) would be a huge step forward.  Personally, I think that the backup jobs should be flushed to an archive regularly enough where rolling over-writes (similar to recycling tapes) can happen to reuse space for more recent backup data.  If it is important for posterity, archiving is the way to go.

Not expandable

The v3600 is not expandable.  The out-of-box configuration is as good as it gets, which is a little dissapointing given that the appliance is so straightforward.  While looking at the demo of this applicance, I also came across a Dell offering powered by Backup Exec that seemed a bit larger than the v3600.

Tape Support

I know that many organizations are moving (or have moved) away from tape and even many who attended Tech Field Day with me were quite vocal about getting away from tape, but what about as an archival solution?

While I also know that backup is not a legitimate archival solution, bringing some archival to the appliance, to allow for backups to have a scheduled archival from the v3600 to a tape library or drive for easy off site archival of data would be a great addition to this already interesting appliance.

Central Administration… not included

For those of you using many Symantec Backup Exec media servers and central administration to manage all of these, this license is not included with the v3600 and will need to be purchased separately.  Might still be worth considering if there are going to be other media servers within your environment.

I would like to see a slimmed down version of the CA server on the appliance so that it can be the one stop shop and manage all of your other media servers (if any).  Maybe the device detects other Symantec Backup Exec servers in an environment and prompts to enable the central administration features if needed.

How does dedupe happen?

Looking back at deduplication which is a nice feature given the finite amount of storage included, the process for dedupe is to point the backup jobs needing to be deduplicated at a predetermined folder on the device.  When jobs use this folder, the data will be deduplicated.  Thats a pretty cool idea.  The use case I can think of for this would be an mailbox backup job.  Duplicate messages would be taken care of automatically.  Hopefully when used with email or other common sources of duplicate files, the rehydration of restored data is optional or at least configurable to allow dupes to be skipped going back to Exchange.

Support and Maintenance

There will be 1 year and 3 year options for support available as a separate purchase.  At first I wasn’t sure this would work, having the support separate, but thinking about it further it makes sense to allow support to be optional.  This way it can be used when needed.  Suppose you have the appliance for a few years with a support contract, if other things come up and the solution no longer fits or will be used only for a certain situation, the support can be changed to a lesser plan (if you have the 3 year option) or discontinued as needed.

Final Thoughts

I am definitely interested in seeing pricing and a bit more about the Symantec Backup Exec appliance.  When that happens I will consider it to possibly become a part of disaster recovery planning.  I do not think at this time it will completely replace the media server we have today, but when my organization goes virtual the data portion, file servers and things that users cannot live without might be great candidates for an all in one appliance.  I will definitely keep it in mind.

My plans for reviewing/playing around in Windows Storage Server 2008 R2 have been quite up in the air lately, but may finally be coming to fruition, although differently than I originally planned.  Since I have installed an EMC Celerra as my organizations primary storage, I thought testing a WSS front end to that system might be appropriate.  In a previous post, I was planning to test this in my home lab, but given the considerations I am looking at for my organization, testing there makes loads more sense.

Initially, I was going to create CIFS shares on the EMC and push those out to the users, purely for the ease of use that comes with CIFS.  Then I got to thinking, in our previous/current environment the WSS 2003 server is providing much the same access to files and other tools allow for the monitoring of files and folders on the shares.

With this additional monitoring capability, which we already own, and more robust storage on the back end, maybe Windows Storage Server 2008 R2 is something to consider.  The only caveat here, at least for me, is that once the testing is over and the pony up phase begins, I will not reduce the footprint in the closet because there will need to be an appliance running Storage Server.

Since I am going to test things first using a virtual machine, I will leave the worrying about footprint and power and all of that until I get closer to knowing what makes the most sense.  In a decent sized virtual environment, with the right hosts I do think a virtual appliance version of WSS would be a nice thing to consider, but we’ll see how the testing goes.