Compellent C-drive Conference Overall reflections

13 05 2010

Here are some of the key things that I learned at the Compellent C-drive conference. It was a very helpful time for me to go and learn about how our Compellent system works, the new features that were implemented in our newest code upgrade that we can start utilizing, and storage and the industry in general.

Below are some of the key things that were of interest to me.

New features we can utilize:
Volume create and edit options expanded:
* Can create volumes in EM and expand volumes (if you “register” them in EM)
* Can format VM’s and map to ESX server from EM

Ability to create clusters of machines, so as to not need to map volumes to individual servers multiple times
Can put a group of volumes in a consistency group, and it will take replays at the same time of all volumes in the group (useful for databases and its logs)
Using Powershell scripts, we can automate vm creations, creation of volumes, and mapping to servers (Compellent has their own commands to interface with powershell)
Chargeback report can be utilized in EM to track how much storage and I/O a department is using, and put a dollar amount to the usage automatically
Green Savings report can be used in EM to see how much money having a Compellent SAN is actually saving us compared to a traditional SAN environment

Attended various vendor and Compellent user seminars:
SpectraLogic: Tape technologies are reliable- reliability has increased 700% over the technology available a decade earlier
Appleby: Riverbed compression product used to replicate data to DR site
Latisys: How to use Compellent and VMWare for effective DR

Interesting facts acquired:
File storage and unstructured data is what’s growing the most, so there will be a large growth in these types of applications
Compellent RAID is self-tuning: if pages move from the tiers to another tier, it will restripe (if too much is moved) to make sure that it is “safe”
Data progression is critical to a web hosting business, as I/O is the secret killer of managed storage practices
The fluid truth: When a better idea comes along, the old ideas tend not to welcome it with open arms. Compellent is fluid because it allows data to change where it’s located dynamically.
Since 2004, Compellent has grown 30%, while EMC and NetApp have both declined, 6% decline, and 3% decline, respectively.
Compellent has over 3000 systems installed in 35 different countries and over 16 countries represented at C-drive (about 600 people at the conference).
Compellent has a patent pending on RAID 6 technique
68% of Compelent customers replicate their data to another site

Personal learning:
Better understanding of RAID, hardware used in SAN (HBAs, enclosures, controllers; difference between Fibre Channel disks, SATA disks)
Understanding of how Data Progression works
Redundancy (RAID levels) are set at the specific tiers)
Compellent RAID is self-tuning: if pages move from the tiers to another tier, it will restripe (if too much is moved) to make sure that it is “safe”
Was introduced to Riverbed, which is a software that compresses data as it travels from one data center to another (for replication purposes)

I was able to network and talk with a lot of different people. Most of the businesses represented were mid-size, and had 2 Compellent SANs (one for replication). There were a few people that I talked with who’s companies had 8-10 Compellent SANs! Having a second Compellent SAN for replication (for DR) was something that most people either had, or were looking in to, as their data is crucial to the running of their businesses.

The best seminar that I went to was How Data Instant Replay and Data Progression Work Together, and it was about “A day in the life of a page”. The presenter took you through the data being created, stored in RAID 10 initially, moved to RAID 5 on Tier 1 still after a replay is taken, and then moved down to Tier 3, RAID 10DM or RAID 6. It really helped cement in my mind how data gets moved, and why it’s so important that Compellent’s system does this. The ability for the Compellent SAN to do this is what makes it so useful and cost-saving. It allows the data you don’t use to move on down to cheaper storage, where it will take a longer time to access if you need to pull it back up, but since it probably won’t be used, its place on cheaper, slower storage is fine.

Some of the fun you missed out on if you didn’t attend, which hopefully the rest of our team will get to have next year:
Smashing of the SAN: – Yes, this was real.
Skit of Storage Heroes: (This is kind of long, but you could get the gist of it by watching a clip)

Someone stated to me that they were impressed that my company would be willing to send me, someone so young and new to the team, to a conference by myself. They said that it showed a lot about the University’s belief in my competence and the willingness of the university to support their staff in training. With that said, I’m thankful that Biola University was willing to let me attend the C-drive conference. It was very helpful for me, professionally and personally, to be able to attend this conference, network with others in the field, learn more about storage.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: