Edward M. Goldberg

Cloud Computing - News and Ideas

Edward M. Goldberg header image 4

myCloudWatcher provides Managed Cloud Hosting today

October 4th, 2010 by Edward M. Goldberg
Respond

myCloudWatcher Provides Managed Cloud Hosting Today

We provide Cloud Deployment support that transitions into 24×7x365 Monitoring, Alerts, and Escalations for the life cycle of your projects.

myCloudWatcher implements “Best Practice” back-up solutions for your projects, including on-going challenges to your back-up systems.

Our team of Cloud Computing veterans, Cloud Specialists and DBA’s are available by phone. We manage your outside services such as ProjectLocker, New Relic, DNS Made Easy, GoDaddy. We provide convenient consolidated billing of cloud services, including DNS, CERTs, AWS, RackSpace, Monitoring, and API; we also manage your renewals. myCloudWatcher has the advantage of seeing the broad landscape: millions of users, thousands of servers, hundreds of projects….Visualize the trend.

Services Provided

  • GoDaddy

- Small low end servers for low cost needs.

  • RackSpace - CloudServers, CloudFiles, CDN

- Great for a fall back server farm.

  • AWS - EC2, RDS, Database Servers, DBShards

- The servers in the main farm.

  • Various ISP Servers

Life Cycle

  • New product updates (code push)
  • Back-ups
  • Server Rotation
  • Auto-scaling
  • Disaster Recovery

Alerts, Monitoring, and Escalations

  • myCloudWatcher continuously tunes all alerts to you application needs to provide early warning and trend analysis
  • Our detailed trend analysis of your deployments are road-maps for 100% availability
  • Our high level view of the internet allows us to distinguish service level problems from individual deployment problems. “Is it Facebook’s issue or is it my game?”

Implementation of Best Practice for Back Up Solutions

  • Back Up strategy challenged to test for accessibility and integrity
  • myCloudWatcher goes beyond the quick-fix to root cause analysis, focusing your engineers to quickly resolve the issues
  • Detailed root cause analysis translates to fast and precise disaster recovery

What you need

  • A full customer support person that is just a phone call away.
  • Direct contact with cloud specialists that are cloud computing veterans, including DBA’s
  • A person to call that can see The Big Picture for advice and Disaster recovery help.

One-stop Shopping

  • Management of Outside services such as New Relic, DNS Made Easy, GoDaddy, and more
  • We keep track of your many subscription renewals
  • Consolidated billing: One bill for DNS, CERTs, AWS, RackSpace, and an aggregation of monitoring systems.

Tags:   · · · · · No Comments.

How to Leverage Amazon AWS RDS today

November 3rd, 2009 by Edward M. Goldberg
Respond

Today, AWS Amazon RDS (Relational Database Servers in the AWS Cloud) are a new Beta and needs to be used as a “peek at that the future” for the IT staff and CTO.  Developers need to wrap their minds around Relational Database as a service.  The time has come for this to be a real SaaS.

The “Big Table” was easy to “pu pu” and say “It does not fit the model”.  RDS is the “real thing” a transactional Relational Database with all the bells and …

IMHO,  we need to start looking at this new model by placing non-mission critical Backup Systems for Fail Over in RDS.  The cost of transfers into RDS is very small.  The cost for running 1 hour a day is cents.  When you need to fail over,  you can launch a HUGE db server and get ton of performance to make up for the remote location if you are “outside” EC2.  If you are “inside” ec2 this is a no-brainier.

1) First Step,  Create an RDS server with RS use a small instance.
2) Each day copy over the “backup” to a live RDS instance.
3) Turn it off.  The RDS makes a Final-SNAP backup on Deletion (read Termination)

On Fail:

1)  Start the RDS on a large server as needed for production, from back SNAP.
2)  You are up and running

COST:

Eleven (11) cents a day!!!!!  ( I assume you use a SMALL Right?)

Save:

You are up and working with a real MySQL 5.1 in under 15 min.  (Worst case) when your DB crumbles and dies one day.

Side effects:

Now you can use the RDS server each day to:

1) Test that your backups work,  NICE….

2) Do a roll up of the “stats” that the DBA was telling you “takes the site down,  stop that!!” (just use the RDS off line, VERY NICE..)

3) Test out new database ideas at now risk to production.  Or two ideas at the same time,  I love that clone feature…
4) The list of off line uses for the DB backup copy go on and on….

Tags:   · · · No Comments.

AWS RDS Rocks!!!

October 31st, 2009 by Edward M. Goldberg
Respond

This time AWS does it right. The Relational Database Servers in AWS called “RDS” are the “Real Thing”.

I just transferred over my whole Drupal DB in a few moments to an RDS Instance. It was very simple.

Let me show you the one-line command solution:

(Warning Linux command comes next!)

$ mysqldump –add-drop-table –databases Comments EdwardMGoldberg NOC myCloudWatcher_drupal wikidb | mysql -h mycloudwatcher.xxx.us-east-1.rds.amazonaws.com -u TheCloudWatcher -pXxXxXxXxXx4

(all password info is fake)

Let me break down this command into the parts:

mysqldump:

This is the LINUX command line database dump program. I just used the default connection (defined in my.cnf) and dumped a few tables I wanted. I added the drop table option so I can update as many times as I like.

mysql -h :

This does the “remote” connection to the new RDS instance and most of the heavy lifting.

So with one line of shell script, I moved into the Cloud.

The RightScale Dashboard was used to set up my RDS instance. It made the whole process of configuration a point and click process.

Next step, look at the results and check out the tables:

http://dev.mysql.com/downloads/gui-tools/5.0.html

Download these great GUI tools and connect to the Database from your Laptop. You do have to “allow” the IP address for you Laptop in the configuration.

It took me about one hour to get setup and transferred into the Cloud. Not Bad.

What’s next? Get my monitor system to track the RDS and alert me when issues happen. I like collected for that task. so that will be my next step.

Take an hour and set up an RDS server for your database. It works great as a “Cold Backup Server”. You pay only a few cents for data storage on the “Cold RDS Server” stored as SNAP data. When you need to “Fail Over” in the Disaster Recovery (DR) mode just Launch the RDS from SNAP. In a very short amount of time you are back on line. To keep the DB up-to-date just launch it once a day for one hour and update the contents. Keep the “Delta” ready for the last few records you may need to add at Launch.

For DR and fail Over RDS ROCKS!!!!

It may be in Beta for now, but start learning how to use this service in the Cloud and never be caught without a Database Server again.

When the Beta ends you will be ready to use RDS full time for your next project.

If you are in the process of learning SQL, the RDS servers will provide you a real Database to do your homework with.

Tags:   · · · · · No Comments.

New Relic & RightScale - Ruby on Rails Life Cycle Tools

April 5th, 2009 by Edward M. Goldberg
Respond

Cloud Computing and Rails are a natural good fit.  With tools provided by New Relic integrated into the RightScale Deployment and Monitoring platform you have a Best Practice Monitoring system.

I Deploy and Watch servers all day as part of the http://myCloudWatcher.com service.  When I started to build the demo for this next RightScale Webinar I learned more about my application in the first few hours with the tools then I had ever seen before.

Please come join me at the Webinar and learn how you can instrument you next Deployment and provide top quality service with your next project.

The Webinar will be live:

Register here

What: Live Webinar — How to Deploy and Manage Rails Apps on the Cloud
When: Thursday, April 9, 2009 at 11 am PT / 2 pm ET
Who: RightScale & Steve Hudson, Director at New Relic

Or you can watch the movie later on the RightScale WebSite:

http://www.rightscale.com/news_events/webinars.php

The New Relic Tools Web Page:

http://newrelic.com/index.html

Tags: No Comments.

Great new Tools like CloudBerry from CloudBerrylab.com

March 2nd, 2009 by Edward M. Goldberg
Respond

To make files work properly with CloudFront you need to set the Meta-Data and permissions on each file in the source  bucket with great care.  I use CloudBerry to upload and check all of the files in my Buckets on S3.

This new tool makes keeping your files correct with just a few clicks of the mouse.

Take a peek at:

http://cloudberrylab.com/

And read the background information on the BLOG:

http://blog.cloudberrylab.com/

Good work guys at Cloud Berry Labs!!!

Tags: 1 Comment

WaaS - myCloudWatcher.com Comes to life!

March 1st, 2009 by Edward M. Goldberg
Respond

After many years of working on Cloud Computing Projects. I have finally started my own project. The Cloud Watcher project provides Managed Hosting for Cloud Deployment Projects.

So many of my clients have asked me to stick around and watch the servers for them after I finished the New Deployment Process. I enjoy training and Launching new projects. But I find it hard to just walk away and not take care on them after the first “Production Launch” is over.

So myCloudWatcher.com was born. Now I Watch many servers in many Clouds.  I get alerts and take some actions for the deployments.

The most important feature is that I watch out for new features and trends that could impact your project.  You get a team of people that look out for the servers and keep them well utilized and working well.

Process and work flow are very important to me. I feel that SVN Source Control and the Trac Ticketing program are important parts of the life cycle of a Project.

Please come talk to me about your Project needs. I would enjoy watching your servers for you.

Did I just invent WaaS ? Watching as a Service?

http://www.myCloudWatcher.com/

Edward M. Goldberg

Tags:   · · · 1 Comment

RightScale and SPLUNK a great team to have on your side.

February 6th, 2009 by Edward M. Goldberg
Respond

SPLUNK> just added Partner templates and Scripts (RightScrips) to the http://RightScale.com service.

This is a great pair to have on your team.  These two players add depth of monitoring to your Deployments.  RightScale takes the low level server metrics very nicely and SPLUNK> watches all of the application logs for you.

Now with this pair set up on every server in every Cloud Deployment.  You have one place to search the indexed logs for all of your servers and control all of the Deployments in any location from one place.

I will be doing a Webinar talking with Michael Wilde of SPLUNK> for one hour.  We will explain all about these products and show you live how to set up a set of servers with RightScale + SPLUNK> .

This event is by invitation only so use this link to get a “seat”:

https://www2.gotomeeting.com/register/280094898

If you missed the last few Webinar and waant to get ready for this one,  watch these first;

http://www.rightscale.com/news_events/webinars.php

The basic idea is servers watching servers.  With SPLUNK and RightScale watching your servers for you and collecting information you will know exactly what your servers are doing at all times.  This is a great way to prove your SLA was meet.  When you need to know what the performance of the Deployment Solution you created provided you need tools like these.

I think that it is important dispel the idea that Cloud Servers are “out of reach” and hard to examine.  In my work tools like these provide the real in depth information in real time.  Just looking at a server in a server room has little value for me.  I look at server photos in catalogs.  I watch servers perform with good tools like SPLUNK> and the RightScale Dashboard.

Tags:   · · No Comments.

How to you prove you met the performance SLA?

January 26th, 2009 by Edward M. Goldberg
Respond

The way to prove that you meet the SLA requirements is:

By setting up several external servers that watch your servers.

When I set up a deployment in the Cloud I also set up several ways to check and see if the service works well.  The more ways I can find to check and check the servers that check (and test them if I get a chance) the better feeling I get the next time I am asked:

How well are you doing,  does the service meet the SLA?

So What services are around?

I just found a new service called: http://WebCron.Org

They are in France. This is great. One more opinion from a server on the other side of the pond from AWS to check each 5 min. and see if my servers providing the service are “200″ over the internet.  This is a good example of a web service to outsource a task.

When you set up with a little magic from Google E-Mail filters. I get an E-Mail and a phone alert each time the reply from the site is != 200. Simple to set up and cheap.

This is not to say that this is a 100% one stop shopping solution for site up time checks. This is just one more tool to use and get one more opinion from France.

I keep looking for ways to get “The warm fuzzy feeling” that all is well. This just takes me closer to that state.

Good work http://WebCron.Org/ you have a nice product. Any more new stuff on the way? Keep me posted.

Tags:   · · · · · No Comments.

Best Practices in the Cloud Series: Deploying a Major System Upgrade

January 21st, 2009 by Edward M. Goldberg
Respond

I will demonstrate how much more efficient and easier it is to perform maintenance tasks in the cloud. And prove that you stay in control through the entire process.

You will see live demonstrations of two use cases: How to respond to a server failure and how to deploy a major system update in the cloud.

You’ll see that much of what you have to do manually in the data center is automated on the cloud. You can operate at a much higher level and accomplish the same tasks with less time and frustration.

  • What: Live Webinar - Deploying a Major System Upgrade
  • When: Thursday, January 22nd, 2009 at 11:00 am PST, 2:00 pm EST, 20:00 CET
  • Who: RightScale Core Development Team

Register Today

Our Webinar Series, Best Practices in the Cloud, is designed to provide you with the tools and training to make your cloud deployment a success.

We hope you take advantage of this opportunity to talk directly with the RightScale Development and Engineering teams. Click here to register!

Tags:   · · No Comments.

Powered By Cloud 2-3 Feb. In London England

January 20th, 2009 by Edward M. Goldberg
Respond

Breaking News,

RightScale’s very own MICHAEL CRANDELL, CEO

Will be at:
The rise and meaning of Cloud Computing

Talking about:

Technologies & Cloud Computing

Obstacles to roll-out and the innovations that will overcome them

Despite all the promises that Cloud computing can bring, some IT departments are hesitant to move across the rubicon. Concerns over security, control, management and integration stand in the way of cloud computing in the enterprise. Cloud computing represents a key shift in how the application stack and platform is built. How will Cloud overcome and integrate traditional enterprise IT silos of database, applications, networking and storage?

http://www.poweredbycloud.com/speakers/mcrandell.aspx

Researchers & Producers of Powered By Cloud  - 2-3 February 2009, Lewis Media Centre, Millbank, Westminster, London UK

Here is a link to sign up for the show:

http://www.poweredbycloud.com/register/registernow.aspx

So many great events this yes for Cloud Computing.

Tags:   · · · · No Comments.