Protecting Data in an Era of Content Creation – Presented by Softchoice + EMC

Editor's Notes

• #7: Would like the T to just grow.Point – for the past few decades the focus of IT has been on the “T”From the time of the very first computers, their use has been focused on repetitive and structured tasks because of their inability to understand our world. This is why most business applications are focused on structuring data into databases and applications.
• #8: Would like the T to just grow.Point – for the past few decades the focus of IT has been on the “T”From the time of the very first computers, their use has been focused on repetitive and structured tasks because of their inability to understand our world. This is why most business applications are focused on structuring data into databases and applications.
• #9: Humans Collect and ContextualizeDataData Processed By ApplicationsHumans Receive and Interpret Data
• #10: Would like a better representation. This is here for concept only. Would rather have this be a build slide of individual objects – hairspray, hair gel, hair cream, etc.Want to make it a little funny or find a funny clip to get the concept that those in the industry call all of this one thing – PRODUCT.
• #12: Would like the T to just grow.Point – for the past few decades the focus of IT has been on the “T”From the time of the very first computers, their use has been focused on repetitive and structured tasks because of their inability to understand our world. This is why most business applications are focused on structuring data into databases and applications.
• #13: Google #1 search engineYoutube #2 search engine
• #14: Present this concept however we want…I want the last part of the presentation to be about tomorrow “Big Data” and “Content”. Just want to get the idea across that Tomorrow is not an option if we do not do today well. However, we need to make decisions today that set us up well for tomorrow.I like the first phrase above because it is more concise and sets the framework for both messages.
• #15: Virtualization delivers big benefits – in efficiency, productivity and agility. There are 3 simple phases in journey.Phase 1 is about lowering costs, focusing on IT applications that have no cost of downtime.With the savings and confidence you built in phase 1, you can start virtualizing mission critical applications. Phase 2 is much more about improving service levels rather than cost savings. There are still savings to be had, but focus is really on reducing downtime and improving agility.In Phase 3 you’re able to run IT as a service. Complete the transformation by building a service catalog, delivering portals that allow faster time to provision infrastructure and are heavily automated and aligns technology to business value. In phase 3, IT is able to drive greater business agility, investing their time in understanding the business and delivering new products / projects faster.
• #16: But there are obstacles (specifically storage and backup related) you will face as you move to greater virtualization – especially as you move through Phase 1 to Phase 2 – that can slow you down.According to Forrester Research, customers have identified:Performance and Completing backups on time;Efficiently Managing capacity;And Controlling costs;as top challenges related to virtualization. Source: Forrester Research (Reichman, Forrester, May 2011)
• #17: Traditional storage and backup solutions are the reasons those challenges were listed. Why?Traditional storage and backup was built for physical infrastructure – and can derail your virtualization efforts. Traditional storage can’t handle dynamic, flexible virtual environments—so applications slow down.Traditional backup treats Virtual Machines like physical machines – so backups are reduced to a crawl having to move redundant data. Neither provides simple management for a virtual environment – lacking the means to discover and manage this new world. It’s costly as well – over-provisioning storage to speed applications and backing up and storing more data in virtual environments. The ultimate result is stalled virtualization efforts - and further cost increases.
• #18: Roman/Greek – perfected Concrete and Arch/DomesSteam engine first built by Richard Trevithick in early 1800’s. Dominated transportation, West, Cities, IndustrializationElisha Otis – perfected Elevator for safety enabling high rise, urban developmentAs Look to the Cloud – confluence of 3 technologies
• #20: Flash delivers performance levels that can be 1,000,000 better than hard disk drives.There is a relatively small amount of hot data, that goes on Flash where the performance great. The cold data goes on spinning drives that can be compressed, deduped, spun down and otherwise made to be highly efficient.The challenge has been deciding when to use Flash and doing that in a manual way is next to impossible. We have found that we can optimized almost all workloads more effectively using an automated tiering method that we share across our entire portfolio. We call this FAST or Fully Automated Storage Tiering and it’s in our VNX family and our VMAX family.
• #21: Here is a FENOMONAL example.VANDERBILT UNVERSITY. This was using VNX.The university has a unique mix of student and faculty data to keep track on.And as YOU CAN IMAGINE: MUCH OF THIS DATA WAS OLDER – but still needed to be on-lineONLY A SMALL PORTION of data was HIGH ACTIVE at any point in time.VANDERBILT WAS THE PERFECT CANDIDATE for a FLASH 1st STRATEGY.They already had EMC Unified storage deployed and now added (8) 100GB EFD for FAST Cache and and (16) 100GB EFD for FAST VP which was only 1.6% of the on-line dataThat SMALL AMOUT of FLASH now serves 78% of ALL IO.The STUDENTS are HAPPY. The FACULTY is HAPPY. AND IT are THE HERO’s of VANDERBILT!Let’s give them a BIG HAND <APPLAUSE>Yes, FLASH FIRST is a BIG DEAL and it is ROLLING OUT EVERYWHERE<CLIC>
• #24: Again, let’s talk about some of the real-world results. What we’re looking at is a full backup. So these numbers are in comparison to doing a daily full backup. Avamar’s savings are dramatic here. Compared to a traditional backup agent, Avamar:Significantly less CPU impactGive CPU cycles back to the applicationsSignificantly less Network utilizationDon’t be a backup scheduling masterSignificantly less Disk I/OGive back to the applicationAt the end of the day your production applications should control the size of your Virtual infrastructureBackup is not ‘strategic’, it is a necessity – don’t let it control your Virtual infrastructure Finally, the ability to control the amount of resources available the Avamar agent (via a throttle) ensures consistent performance for applications during backup.Note to Presenter: Lab layout for full backup comparison, ESX Server:Dell 6850: 8 CPU, 64 GB RAM, 4 virtual machinesVirtual Machines: traditional backup client installed, Avamar client installed, Windows 2003 SP2Avamar: Avamar Virtual Edition for VMware 3.7.2, RAID 5 CX storage (single 800 GB LUN)Backup software: traditional backup software, backup target is RAID 5 CX storage
• #25: Need to edit this clip to include the following – First 6 second intro…then go to question at 0:48 in the presentation and play through the rest of the slide.
• #26: More examples of unstructured data and large scale structured data that will need to be correlated in an Information Centric age.The sources of information are expanding. Many new sources are machine generated. It’s also big files (siesmic scans can be 5TB per file) and massive numbers of small files (email, social media).Not all businesses will deal with some of there, but these are just the applications that we can think of today. There will be many more things to come that we can’t think of today.
• #28: It is no longer about the data, it is about meaning.Where we are going with technology is that we will see applications and systems on a new scale where IT will be more about people and less about machines.People do not live their lives in rows and tables. Unstructured data is data in human friendly form an that is where the interesting stuff happens.
• #29: The idea with Big Data Analytic platforms is that like Google, you can search and access all of your structure and unstructured data sources, analyze this data in real-time, on systems that offer scale out and co-processing of both structured and unstructured data.In essence this is what EMC Greenplum with its database for structured data and Hadoop platform for unstructured data.
• #31: It would be great if there was an image here to convey this message.Key thoughts: We will continue to do the same business / structured functions we do today. Some will be able to use information from this new world as appropriate.It will be an incremental step in getting there. Function will come before scale – this is why the push to unified and virtualized storage is so important.Some systems – Data Protection, for example, will be leveraged by both worlds.