ProductionScale RSS Feedhttp://www.productionscale.com/home/Scalable Systems Architectureen-US2012-02-05T08:50:52ZSquarespace Site Server v5.11.81 (http://www.squarespace.com/)People, Process & Technologyhttp://www.productionscale.com/home/2012/1/31/people-process-technology.htmlKent Langley2012-02-01T03:15:02ZI am often asked how I do things and why. In most cases it really does come down to thinking about three things and making sure they are properly tended to at all times. Those three things are people, process and technology in that order.

With the right people you can do anything you can set your mind to doing within the realms of possibility. With the right process you will be far more effective than the next team and your operations will be able to scale. With the right technology you will be enough out on the edge to get ahead but rock solid and scalable so that you will not crumble under brittleness or technical debt (also related to process mind you).

I haven’t been blogging much lately because I’m really quite wrapped up in my work for my clients as well as just trying to spend more time with my family. I am trying to find ways to get back into the blogging grove more than once a month though. I’ve been building Social CRM and Business Analytics solutions for some clients on two interesting technology stacks. One stack is a Riak, Redis, RabbitMQ, Jetty, Scala/Java, Akka, Nginx core stack on AWS. The other is MongoDB, RabbitMQ, PHP/Symfony2, Node.js, Nginx core stack on AWS. There are other techs involved of course. They both use Chef extensively for example. One uses CI with Jenkins as part of the process, both use automated deployment tools in Ruby (capistrano primarily). Both of these are slated for public release in short order so I’ll be able to say more then. It’s been truly interesting working on them in parallel and I’m very excited to see them go live over the next weeks.

In both cases, it’s the team, the process, and the technology that have made it possible to process large amounts of data, deploy tremendous amounts of code fast, and create products that have excellent potential to make a difference to their users and scale well over time.

So, if you ask me to build you something you’ll be hearing me ask if you are willing to make the commitment to the people, processes, and technologies that will make it a reality fast. When it all comes together it is great fun!

I hope to get some time to write more about the actual details of this trinity in the near future. Be patient, I have to pay the bills you know!]]>How NOT to Sell NoSQL Databasehttp://www.productionscale.com/home/2011/12/21/how-not-to-sell-nosql-database.htmlKent Langley2011-12-21T17:01:26ZThis is the description of my first experience with a newer NoSQL database that we'll just call NoSQL Database #9999 I was told about and asked what I thought about it overall. I hadn't heard of it before but I wanted to see what the deal was since I work with several others. I'm always up to see if something is actually the new hotness.

I found marketecture diagrams everywhere. The development cycle is closed and opaque development for server and client. There is a 30 day "free trial" signup wall to maybe get to the download screen. I'm not sure since I didn't fill it out and I really don't feel like spending my time navigating a sales channel for filling it out. The License agreement was really fun. The short version is that it is a non-exclusive licensing model and no ability to use/test in production to see if it really works. The choice parts basically say that I can't use the software in production and that says that if it doesn't work that's not our problem and we never said it would. There is actually a warranty clause that says they don't warranty anything at all it's is just "as-is" without warranty! I am not feeling the love at this point. Then, I wanted to see the pricing. I couldn't of course. The minimum contract term beyond 1st 30 days is 12mos with, you guessed it, unspecified pricing information unless I contact sales directly.

So, now I know why I've never heard of this software and nothing meaningful has been written about that is not PR or Marketing driven. There is really no way that I would even consider adopting this software at this point. It's most likely that it is not real.

So, NoSQL database #9999 there are many other equally usable solutions that are far more transparent in the way they do business and foster community around their products. This isn't about paying money. This is about trust. So, sorry NoSQL #9999, but I'll not be entering your sales cycle in this fashion or evaluating your product at this time. Moving along now...

Happy Wednesday Everyone!

]]>Building an Application upon Riak - Part 1http://www.productionscale.com/home/2011/11/20/building-an-application-upon-riak-part-1.htmlKent Langley2011-11-21T01:13:16ZInteresting Things business cloud cloud computing cloud native application distributed distributed nosql riakFor the past few months some of my colleagues and I have been developing an application with Riak as the primary persistent data store. This has been a very interesting journey from beginning to now. I wanted to take a few minute and write a quick "off the top of my head" post about some of the things we learned along the way. In writing this I realized that our journey breaks down into a handful of categories:

Making the Decision
Learning
Operating
Scaling
Mistakes

We made the decision to use Riak around January of 2011 for our application. We looked at HBase, Cassandra, Riak, MySQL, Postgres, MongoDB, Oracle, and a few others. There were a lot of things we didn’t know about our application back then. This is a very important point.

In any event, I’ll not bore you with all the details but we chose Riak. We originally chose it because we felt it would be easy to manage as our data volume grew as well as because published benchmarks looked very promising, we wanted something based on the dynamo model, adjustable CAP properties per “bucket”, speed, our “schema”, data volume capacity plan, data model, and a few other things.

Some of the Stack Details

The primary programming language for our project is Scala. There is no reasonable scala client at the moment that is kept up to date for Riak so we use the Java client.

We are running our application (a rather interesting business analytics platform if I do say so myself) on AWS using Ubuntu images.

We do all of our configuration management, cloud instance management, monitoring harnesses, maintenance, EC2 instance management, and much more with Opscode Chef. But, that’s a whole other story.

We are currently running Riak 1.0.1 and will get to 1.0.2 soon. We started on 0.12.0 I think it was... maybe 0.13.0. I’ll have to go back and check.

On to some of the learning (and mistakes)

Up and Running - Getting started with Riak is very easy, very affordable, and covered well in the documentation. Honestly, it couldn't be much easier. But then... things get a bit more interesting.

REST ye not - Riak allows you to use a REST API over HTTP to interact with the data store. This is really nice for getting started. It’s really slow for actually building your applications. This was one of the first easy buttons we de-commissioned. We had to move to the protocol buffers interface for everything. In hind sight this makes sense but we really did originally expect to get more out of the REST interface. It was completely not usable in our case.

Balancing the Load - Riak doesn’t do much for you when it comes to load balancing your various types of requests. We settled, courtesy of our crafty operations team on an on application node haproxy to shuttle requests to and from the various nodes. Let me warn you. This has worked for us but there be demons here! The configuration details of running HA proxy to Riak are about as clear as mud and there isn’t much help to be found at the moment. This was one of those moments over time that I really wished for the client to be a bit smarter.

Now, when nodes start dying, getting to busy, or whatever might come up you’ll be relying on your proxy (haproxy or otherwise) to handle this for you. We don’t consider ourselves done at all on this point but we’ll get there.

Link Walking (err.. Ambling) - We modeled much of our early data relationships using link walking. The learning? S-L-O-W. Had to remove it completely. Play with it but don’t plan on using this in production out of the gate. I think there is much potential here and we’ll be returning to this feature for some less latency sensitive work I perhaps. Time will tell...

Watchoo Lookin’ for?! Riak Search - When we stared search was a separate project. But, we knew we would have a use for search in our application. So, we did everything we could to plan ahead for that fact. But, by the time we were really getting all hot and heavy (post 1.0.0 deployment) we were finding our a few very interesting things about search. It's VERY slow when you have a large result set. It's just the nature of the way it's implemented. If you think your search result set will return > 2000 items then think long and hard about using Riak's search functions for your primary search. This is, again, one of those things we’ve pulled back on quite a bit. But, the most important bits of learning were to:

Keep Results Sets small
Use Inline fields (this helped us a lot)
Realize that searches run on ONE physical node and one vnode and WILL block (we didn’t really feel this until data really started growing from 100’s of 1000’s of “facets” to millions.

At this point, we are doing everything that we can to minimize the use of search in our application and where we do use it we’re limiting the result sets in various ways and using inline fields pretty successfully. In any event, just remember Riak Search (stand alone or bundled post 1.0.0 is NOT a high performance search engine). Again, this seems obvious now but we did design around a bit and had higher hopes.

OMG It’s broken what’s wrong - The error codes in the early version of Riak we used were useless to us and because we did not start w/ an enterprise support contract it was difficult sometimes to get help. Thankfully, this has improved a lot over time.

Mailing List / IRC dosey-do - Dust off your IRC client and sub to the mailing list. They are great and the Basho Team takes responding there very seriously. We got help countless times this way. Thanks team Basho!

I/O - It’s not easy to run Riak on AWS. It loves I/O. To be fair, they say this loud and clear so that’s my problem. We originally tried fancy EBS setup to speed it up and make it persistent. In the end we ditched all that and went ephemeral. It was dramatically more stable for us overall.

Search Indexes (aka Pain) - Want to re-index? Dump your data and reload. Ouch. Enough said. We are working around this in a variety of ways but I have to believe this will change.

Basho Enterprise Support - Awesome. These guys know their shit. Once you become an enterprise customer they work very hard to help you. For a real world production application you want Enterprise support via the licensing model. Thanks again Basho!

The learning curve - It is a significant change for people to think in an eventually consistent distributed key value or distributed async application terms. Having Riak under the hood means you NEED to think this way. It requires a shifted mindset that, frankly, not a lot of people have today. Build this fact into your dev cycle time or prepare to spend a lot of late nights.

Epiphany - One of the developers at work recently had an epiphany (or maybe we all had a group epiphany). Riak is a distributed key value data store. It is a VERY good one. It’s not a search engine. It’s not a relational database. It’s not a graph database. Etc.. etc.. Let me repeat. Riak is an EXCELLENT distributed key value data store. Use it as such. Since we all had this revelation and adjusted things to take advantage of the fact life has been increasingly nice day by day. Performance is up. Throughput is up. Things are scaling as expected.

In Summary - Reading back through this I felt it came off a bit negative. That's not really fair though. We're talking about nearly a year of learning. I love Riak overall and I would definitely use it again. It's not easy and you really need to make sure the context is correct (as with any database). I think team Basho is just getting started but are off to a very strong start indeed. I still believe Riak will really show it's stripes as we started to scale the application. We have an excellent foundation upon which to build and our application is currently humming along and growing nicely.

I could not have even come close to getting where we are right now with the app we are working on without a good team as well. You need a good devops-like team to build complex distributed web applications.

Lastly and this is the real summary, Riak is a very good key value data store. The rest it can do is neat but for now, I'd recommend using it as a KV datastore.

I'm pretty open to the fact that even with several months of intense development and near ready product under our belt we also are only scratching the surface.

What I'll talk about next is the stack, the choices we've made for developing a distributed scala based app, and how those choices have played out.

]]>The SaaS Aggregation Benefit Miragehttp://www.productionscale.com/home/2011/10/6/the-saas-aggregation-benefit-mirage.htmlKent Langley2011-10-07T00:47:22Zbusiness cloud computing saasIn this service oriented on-demand world I’ve been running into something again and again lately that I’ve found interesting and a bit annoying.

To start, imagine I’m going to build an application that uses two 3rd party services on-demand. We’ll just call them service A and service B and say each have two features. For this example it does not really matter what the services do.

Service A
Feature A-1
Feature A-2
Service B
Feature B-1
Feature B-2

So, I create my application and it first uses service A do something and it uses Feature A-1 and A-2. Then, with the output of that it uses service B to do something else using feature B-2.

Now, a few months down the line when things are going great I get a call from my account manager at Service A telling me I can now get all the features of service B directly from them included. So, what they are telling me is that my service structure now looks like this:

Service A
Feature A-1
Feature A-2
Feature B-1
Feature B-2
Service B
Feature B-1
Feature B-2

On the surface this looks really good. It’s the same thing with less hassle right? Maybe not.

This is where my annoyance surfaces. Dig in and dig in well. What I find again and again is that it’s simply not true because of what I’ll just call the filter effect. What you really are getting with this new and improved service A is more like.

Service A
Feature A-1
Feature A-2
Feature B-1

Notice that Feature B-2 is missing and that probably no body mentioned it. Or, it’s more like:

Service A
Feature A-1
Feature A-2
Feature C-1
Feature C-2
Feature C-3
Feature C-n-OMG
Service B
Feature B-1
Feature B-2

And you don’t care because C isn’t B and all you need as A-1, A-2, and B-2. While they say it’s equal is not and the app use feature B-2 if you’ll recall. How much time did you just spend?

So, by the time you get through all this and figure out that the new improved Service A + B is pretty useless and all you really want is what you already have you will have wasted a lot of time. There are less features, more complexity, less control, and likely much worse service and support for the aggregated services since you have no direct relationship to the end point provider.

So, rambling aside the point is that these service provider mashup aggregaters are not what they often seem on the surface and I’m frequently finding that the best deal is going right to the source and that any “savings” on the surface likely gets eaten up later in a variety of ways that are difficult to predict. In most cases, it’s best to go to the source to get what you want.

]]>Brick and Mortal Retail Doomed, Doomed I Sayhttp://www.productionscale.com/home/2011/9/9/brick-and-mortal-retail-doomed-doomed-i-say.htmlKent Langley2011-09-10T05:34:24ZfutureNot my usual blog topic, but hey, it’s Friday and I had a brutal week. But, I just had to relay a retail experience I had a week or so ago. I went into a local hardware store. It’s a pretty good one but I’ve always thought they were a bit pricey. But, I needed a new vanity mirror and I needed it now. On the display they had one I immediately liked and was willing to pay the premium. Once I found a salesperson I actually started w/ a question about some sconce lights I liked all so. He looks at me. Looks me up and down and says, “your not going to like it.” I said, “hit me.” He did, I didn’t. Whatever. Bad start for sure. In any event, I say, I’d like to get this mirror you have over here on the wall. I took him to it, I said I’d like one of these please. Here are the two things he said to me:

1. We don’t have those in stock.
2. But, I bought one and it’s still in the box at home. Want to make me an offer I might sell it.

Ohhhhkay... I said, alright, I think I’ll stick to the store here. How long to get one.... Turns out it’s 7-10 business days. Keep in mind this is a premium price. Apparently there are six with their nearby suppliers. At this point I had already snapped a picture with my phone and sent it to my wife for approval. She says... LOVE IT... in response. So, I take the product #. I tap it into a google search. I find it on Amazon.com from an affiliate for 30% less, 5-7 days delivered, and a bit of tax. Net savings over store of around 20%. I decide to just show they guy and I said the following things:

1. I’d prefer to buy local. Can you sell it to me for this price and get it here in about the same time frame.
2. I can just order it from here an it’ll come to my doorstep by pushing this buy button now.

The response floored me. No. He mentioned the one he had at home again. I said okay. I hit the buy button and went home.

Here’s the crazy part. This is an employee owned store!

I’ve tested the I’d like to buy this at several other stores recently. Most of them simply have NO stock. They are just display stores. You cannot buy what they have and go home w/ it.

Astounding. Physical retail has absolutely no chance with me using this kind of approach. I’ll just order from home. Shame, I wanted to buy local. Either way mirror will be here in a couple of days along w/ the sconce lights, towel rack, and TP holder to match. *sigh*

]]>Stop Staring at my Polyglot!http://www.productionscale.com/home/2011/9/4/stop-staring-at-my-polyglot.htmlKent Langley2011-09-04T20:40:56Zcloud computing cloud computing cloud native application database distributedI received an interesting comment/question via my blog recently. It went a bit like this...

I’m developing a distributed cloud application but my developers are pushing back on me for having a polyglot database strategy. What should I do?

I won’t get into exactly what it is they are doing since that would take several more pages. This is something of a stream of thought post so apologies if it is a little rough around the edges. The easiest way to answer is in the context of an application I’ve been working on for a while that has some similarities to what this person wants to build. Everything I’m describing is part of an app I’ve been building with a client since earlier this year.

Typically you'll need a few layers of "data storage” for any distributed batch or real time application (cloud native application) which is what I understand that you are trying to build.

I consider anything that holds data that is for presentation, computation, or transformation part of the data storage architecture and I like to break it down by time in storage from least amount of time to most.

Short or Very Short Term: Single node caches (like memcached) or volatile computer node memory
Mid Term: Queue's and IMDG's
Long Term Durable Storage: Dynamo and BigTable derivatives abound

There are numerous database products that live in or even between those tiers these days; more than ever before. By no means is what follows even close to an exhaustive list. A quick list of the ones I have worked with in the last few months personally looks like:

Short Term: Memcached, Redis, RabbitMQ, ZeroMQ, DRAM, APC, MongoDB
Mid-Term: Redis, GridGain, RabbitMQ, ZeroMQ, MongoDB
Long-Term: Riak, MongoDB, S3, Ceph, Swift, CloudFiles, EBS, HBase

Short-Term storage is ALL in memory, not persisted to disk, and not intended to be used for long periods of time. Your application also has to be able to deal with the fact that this type of storage is essentially ephemeral. If the node gets a KILL signal from some source or another your app needs to know how to deal with this gracefully. In other words, storage here is not durable at all.

Mid-Term storage is used for longer running processes. It benefits greatly from being distributed and having a higher degree of durability. This is generally still where most of the work in done in main system memory (no disk I/O) but also where you might do complex calculations or data transformations on your way to your goal. You do it here because it’s fast. You do things here because they can be shared amongst lots of workers (like queue subscribers).

Long-Term storage is used for exactly that, long term durable storage of important data that provides sufficient and reasonable interfaces from which to retrieve that data again when needed. Preferably it’s possible to do things like map-reduce jobs so that you can iterate and retrieve what is necessary which you may then operate on at one of the higher levels up this stack.

You’ll see that I’ve put some of them in all or multiple categories which might seem odd until you understand how they work and match the technology to what ever you are trying to achieve from a business perspective.

I have a tendency to avoid things that require overly complex operational management issues for starting up projects because I like to try to get my TCO (Total Cost of Ownership) over time (3-5 years) as low as possible while achieving the project goals and SLA’s. There are a couple of exceptions on the list above that do have more operational overhead (MongoDB and HBase) but they are good enough in the right context that you might want to learn and use them anyway.

Now, back to the question at hand. Should I use one type of DB or many for the needs at hand. In this case, I’ve told them that they should use as few as possible, possibly only one. The reason for this was that in their case they will value speed, consistency, and lower cost of operations at this early stage of their project. They are developing an interesting distributed system for cool reasons. I recommended a choice to them that I think will help them get to their goals fast and cost effectively while allowing them down the line to break off pieces of the application later as and if needed.

Parting words are that it will, over time, be nearly unavoidable that this (and most) applications of a distributed nature end up being database polyglotoumous. However, I do think it adds a lot of complexity and overhead and in the early stages of a project it's not usually necessary unless what you are doing is of great complexity in which case you might want to break that down anyway to something more manageable.

]]>Can New Clouds Teach Old Apps New Tricks?http://www.productionscale.com/home/2011/8/7/can-new-clouds-teach-old-apps-new-tricks.htmlKent Langley2011-08-08T00:27:06Zcloud cloud native applicationCramming the same old code, CMS, application, etc into the cloud (any cloud) doesn't make the most of the capabilities of cloud computing in all it's various forms. I expect to be discussion this subject more in the near future. But, start by giving two examples and labeling them cloud native application design pattern and anti-pattern.

A Cloud Native Application Design Anti-Pattern

I'll pick on Drupal a bit (but with love). If one installs Drupal at a cloud IaaS or PaaS provider then that does not make Drupal a cloud native application. To me, this seems obvious but I am not so sure it is obvious in general. The Drupal CMS is not a Cloud Native Application. Putting Drupal, Wordpress, CMS XYZ of your choice on cloud computing IaaS or even PaaS provider of your choice essentially means you end up with an virtualized n-tier application running in the cloud with many of the same limitations of a hardware based deployment and only some of the benefits of being a cloud native application running on a cloud computer. Yes, of course, and admirably (see billions of pageviews per month) drupal can run IN the cloud. But, that does not make it OF the cloud. But, I will say that based on personal experience even considering all this situation it's still likely the right choice in a great many cases to run it in the cloud.

A Cloud Native Application Design Pattern

If you want to see what CMS can look like as a cloud native application then check out the Lily CMS project. I personally might not choose this specific architecture and systems design to achieve the same goals. However, there is more than one way to build a CNA. They have done some great work there and are clearly on the right track! It's excellent work and I have respect for what the Outerthought team has created with their platform. It's actually potentially quite a lot more than just a CMS as well. In any event, I think that with the exception of the default HBase high availability limitations (which will be addressed soon by HBase project I suspect) this can be considered a cloud native application. Coupled with the appropriate monitoring, automation, and even cloud environment awareness it would be a very powerful cloud native application.

All of this summarizes to me as one very simple fact. There is a tremendous opportunity ahead! Exciting times.

]]>The NIST Definition of Cloud Computing(Draft)http://www.productionscale.com/home/2011/8/7/the-nist-definition-of-cloud-computingdraft.htmlKent Langley2011-08-07T16:07:11Zcloud cloud computing

click for original doc

I thought I'd start the week with a reminder of an oldie but goodie. This document came out after the intial barrage of "what is cloud computing" and "cloud computing defined" posts from a few years ago. But, I've always felt that NIST did a great job with it overall.

In my opinion it's still one of the best and most complete current definitions of Cloud Computing of any other out there. So, in the off chance that you have not seen this defintion of cloud computing it is definately worth the time to read through.

One of my earliest definition articles from April 2008, Get Your Head in the Clouds, is still my most trafficked article on this site most weeks.

"Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model promotes availability and is composed of five essential characteristics, three service models, and four deployment models..." --NIST Cloud Definition

In particular I like the way they break down the characteristics, service, and deployment models.

]]>Cloud Operating Systems: Do They Exist Yet?http://www.productionscale.com/home/2011/8/6/cloud-operating-systems-do-they-exist-yet.htmlKent Langley2011-08-06T21:10:50Zcloud computing cloud computingI was asked an interesting question by friend of mine a few days ago. He says simply, “Are there any Cloud Operating Systems?”

I then proceeded to argue that no, there really were not any because all of the necessary pre-requisites for being both cloud native while providing the basic services an operating system should provide were not available in the current crop of software claiming to be a cloud operating system upon which I could build cloud native applications.

Let’s think about what an Operating System gives you. An OS is essentially an abstraction layer over hardware that provides the necessary interfaces(drivers) to get the work done that is needed by computer programs. At its most basic an operating system provides storage, networking, and compute and the management of those things in a fairly transparent way. It also provides a tremendous amount of flexibility to it’s user.

Wikipedia defines an Operating System as, “software, consisting of programs and data, that runs on computers, manages computer hardware resources, and provides common services for execution of various application software. The operating system is the most important type of system software in a computer system. Without an operating system, a user cannot run an application program on their computer, unless the application program is self booting.” -wikipedia

If I was to write the definition for wikipedia using the definition for an operation system as the bse then I would say, “A cloud operating system (COS) is software, consisting of programs and data, that runs on IaaS providers, manages IaaS resources, and provides common services for execution of various application software and SaaS. The cloud operating system is the most important type of system software in a cloud computing system. Without a cloud operating system, a user cannot run a cloud native application program on the cloud computer, unless the application program is self booting.

To me, this sounds most like what we currently often call Platform as a Service (PaaS).

The problem I have with current PaaS systems is that they dictate heavily and take away far too much from the programmer from the underlying services. This is a compromise of course. But, a true operating system is far more flexible than the existing crop of PaaS services. Just imagine if your newest operating system upgrade wouldn’t let you access ½ the CPU cores or the networking services for some reason and your manufacture just said, tough cookies because that’s just how it is. You’d probably try to return it if you could.

Where this got interesting for me in thinking about it later was the concept of “self-booting” that was mentioned in the wikipedia defintion above. If you build the necessary intelligence into your PaaS such that it can bootstrap an application, the storage, networking, and compute necessary to run and then manage that moving forward then you do, in fact, have something more resembling a cloud operating system as defined above.

]]>Welcome VGBuilder to the world!http://www.productionscale.com/home/2011/8/2/welcome-vgbuilder-to-the-world.htmlKent Langley2011-08-03T04:53:55Zcloud cloud computingBuild custom apps + deploy them to your cloud host of choice -- straight from the command line.

The first applications using VGBuilder are live already and there are more in works.

This Cloud Native Application Development toolset was created out of a genuine need for a big project about a year ago that needed to move very fast. There were no tools that we felt really met our needs at the time so we created our own. That’s a story that will be told soon in more detail I expect.

I consider VGBuilder a Cloud Native Application development tool set. It does not meet all my requirements as you’ll note if you read the above article. But, in time it will I believe.

To understand some of my thinking that went into VGBuilder it helps to read these articles:

Cloud Native Applications

I wouldn't claim necessarily that VGBuilder is revolutionary but it certainly is evolutionary. One of the more powerful aspects of this toolset is that it removes the need for any middle man or centralized platform to get your ideas from your brain and into the cloud crazy fast and with quite a bit of style. It allows you to build and deploy your applications in the cloud with almost no barriers. It removes the need for PaaS services for many people right from the start.

Essentially your laptop or computer becomes your private cloud and either Amazon or Rackspace (supported today out of the box) becomes your public cloud. This is essentially a hybrid cloud model that puts all the power in the developers hands and automates almost everything that is often tedious or cumbersome otherwise.

Things are still rough around the edges. As much as anything I want to guage interest and get feedback on this new set of tools. There is a long way to go but preliminary case studies have been excellent.

Please contact me or sign up on the early access web page for if you’d like to be kept informed of the progress and provided access to use the tools for your own projects as it is opened it up to more early adopters.

]]>ProductionScale Communication Problem Discoveryhttp://www.productionscale.com/home/2011/7/30/productionscale-communication-problem-discovery.htmlKent Langley2011-07-30T19:38:40ZI discovered today a distressing thing about my blog. For the last few months two things have been misconfigured.

One, my phone number was an old number that is out of service. So, if you tried to call me then please just note the new number on the about page and try again.

Two, my contact submission form was going to an email address that was Supposed to forward the mails but has not being properly performing its duty. I have fixed this issue and gone back through and replied to all the emails that were legitimate.

My sincere apologies for those communications issues and if you have tried to reach me through this blog and I did not respond, that is likely why. I always respond to legitimate requests for information, work, or engaging discourse. So, please send me a new note and don't hesitate to reach out! I thoroughly enjoyed the connections I've made over the years from this blog and hate to think I let it lapse for a bit!

Sincerely,

Kent Langley

]]>Scale Planning and the AKF Scale Cubehttp://www.productionscale.com/home/2011/7/9/scale-planning-and-the-akf-scale-cube.htmlKent Langley2011-07-10T04:39:59Zcloud computing cloud computing distributed distributed scalabilityThere are a lot of ways to draw diagrams for availability and scalability. I use different ones for different purposes all the time. However, when I was reading the Art of Scalability by AKF partners I ran across a nice compound diagram they call the AKF Scale Cube which helps simplify the explanation of the multi-dimensional nature of scalability issues in complex web application scenarios.

I’ve been using this visualization model to help me explain how things fit together in both technical and business discussions. It comes in very handy I must say.

Most recent I’ve been using it to describe a gnarly distributed application I’m working on for a client. What follows is a generalized version of a functional use and some discussion of a compound view of ones I have created for clients of mine.

Some base-line definitions are in order if you haven’t read the book mentioned above.

X-Axis - Horizontal Scalability, Clones, Scale Out. These are terms often associated with the X-Axis. In the case of this graph, day you build a data processor then make 10 copies of it. Well, that’s scaling to 10 units on the x-axis in this graph. Depending on your application, this can help you increase your capacity; but not always!

Y-Axis - I’ve called this axis functional decomposition for a long time. It can be thought of as breaking the application down from a monolithic single instance into discreet stand-alone parts. I have some examples that are a bit of mix from various projects I’ve worked on in the past here.

Z-Axis - This is the tricky one for most folks. This is what people might call sharding, partitioning, etc. Keep in mind, I’m not only talking about a database here. I’m talking about an entire complex multi-faceted distributed highly-available web application.

0,0,0 - The intersection of all three axis or 0,0,0. This is what some would call an all-in-one server. It’s often used for proof of concept for for launching without a care in the world for future growth needs. There’s nothing wrong with it as long as you understand the limitations and technical debt associated with the approach.

Z-Axis Item Explanations

Client - Assuming this is a multi-tentant application you may want to shard your application by client such that each client or group of clients is assigned somehow to a specific cluster of nodes.

Geography - Assuming you’d like to have built in DRBC and your applciation is capable of surviving being split up into many pieces then you could end up sharding your application by data center and broader geographies such as city, state, country.

External Cloud - Using IaaS and PaaS resources outside of your own data centers. For a refresher on IaaS and PaaS see the article I wrote, “Cloud Computing: Get Your Head in the Clouds,” in 2008 that was heavily read over the years.

Internal Cloud - Using your own infrastructure resources BEHIND your own firewall to do whatever it is that your application does. This doesn’t always have to be a cloud. If you want to know what I think it takes to be a cloud then read the several articles I wrote over the years related to that topic. I do set the bar pretty high though I’ve learned.

Purpose - You might want to simply partition along the Z-Axis by any generic purpose for various reasons. I think of this a little bit like saying I want to put all widgets in data node 1 and all waggles in data node 2. They’ll both fit in a single node but maybe I want to spread my risk around. This one is a little nebulous but it can engender fun conversations about why things need to exist at all.

Shard Key - We see this all the time in traditional RDBMS style deployments and even in some of the newer tools in the NOSQL world. It’s basically just some index of what node you put things one somewhere. For those of you that had to deal with libraries before the internet you’ll remember the lovely card catalog. It’s was nicely set up to help you figure out with shard of the library your book was close to. Then, when you got there, good old dewey decimal system kicked in to take you the rest of the way.

Y-Axis Item Explanations

Data Processing - this could be some application that transforms data from one state to another. For example, it might simply remove all the spaces in a document and replace them with dashes. That’s a bit of a silly example, but just to make the point.

Data Aggregator - I’ve had to build project after project that needed one form or another of data aggregation. So, just think of this as something that might consume and RSS feed and stick it in a database of some kind.

Distributed Calculation - I’ve been doing work and research with Map-Reduce, Actor Models, the Bulk Synchronous Parallel Model and more exotic instruments from past, present, and future. This is simply something that does some kind of math or calculation of some sort. For example, counting all the uses of the word onomatopoeia in 50TB of English essays by high school students across 100’s of of compute nodes.

Processor App - This is just a generic discreet application that processes something, like an API request for example.

Web App - This is an application, in my case, written in a modern MVC framework that has the job of interacting with web users and getting things done in the back-ground in various ways with various services.

Base Installation - I think of this as just shared code. One of the developers I have been working with recently suggested that we extract a number of commonly used components from various application pieces on the Y-Axis and build a library of sorts. Great suggestion in this case, so I stuck in on my general diagram to remind me in the future.

What’s interesting about all these conceptual applications is that if you create them correctly and with the correct architectural models that each item that lives on the X Axis will also be able to scale on the X and Z axis. For example, you could have your web application running 5 X-Axis copies in 4 Y-Axis partitions; say external cloud, by client, purpose, and by geography per client. So, you’d end up use four AWS Availability zones in 2 AWS locations running 96 application nodes in total. Of course, your application has to be built correctly to take advantage of all of this distribution at every level. But, that’s a topic for a later date I suppose.

In summary, this post was just to share some of my thinking around the use of a very nice visualization tool by the fine folks at AKF Partners. So, a shout out to them for the nice tool and hopefully this helps people a bit understand how it can be used / thought of in a variety of ways.

Just remember, it's not one-size-fits all. Your use, labels, and needs for such things will vary greatly depending on what you are trying to architect, develop, and deploy.

]]>On Clouds and SPOF’s (or the Great AWS Outage of April 2011)http://www.productionscale.com/home/2011/4/22/on-clouds-and-spofs-or-the-great-aws-outage-of-april-2011.htmlKent Langley2011-04-23T00:58:25ZIT cloud cloud computing distributed
Just a couple of days after posting about cloud native applications Amazon raised the bar by having some issues in one of their data center regions. These issues primarily affected EBS and RDS from what I’ve read. So, pretty much everything one way or another since using AWS EC2 without EBS in any form for most applications that exist today is a little wacky for most folks. This is because your EC2 AMI won’t persist through a reboot in the absense of the use of EBS. Most folks have not reached the operational nirvana yet of full automated configuration management and application fault tolerance that makes this acceptable for them.

What level of SPOF (Single Point of Failure) are you are willing to tolerate. So, I wanted to “scale up” the idea of the SPOF then bring it back down again. Here we go.

If the earth stops working, so will your web application (admittedly there might be some satellite networks that don’t have this problem... but who cares at that point?)

So, let’s keep going. Each of these is a potential single point of failure.

Earth > Continent > Country > State/Region > City > Neighborhood > Building > Floor > Room > Rack > Server > Server Component

And, at each tier, there are numerous dependencies and contexts to keep your service running at any given time. There are the obviously ones like the above example where if the earth explodes the neighborhood is pretty much shot to hell also. But, that’s obvious. It’s gets less obvious when you dig deeper into the data center and see that there are 5 servers so that’s okay right? Maybe. Maybe not. If it is something like.

Dynamic Name Service > Load Balancer > Web Server > Application Server > Database Server

Then those 5 servers/services might be in that one rack per data center per room per building per neighborhood per city per state per country per continent per planet is looking pretty vulnerable. In the grand scheme of things the loss of one power supply in one machine could impact the entire planet’s capacity to retrieve whatever is on that DB that is so globally important; like a picture of your kid making a funny face on his 2nd birthday.

Do you think it is Amazon AWS’s fault if you put that database on one server in one rack in one place with no reasonable SLA and it goes away forever? Not so much. You are accountable and responsible. You made that choice.

Now, how can we change this for the better? We can develop applications that are able to tolerate the loss of a single point of failure at a sufficient granuality (Earth is a bit extreme today) such that our applications keep running when bad things like the AWS outage occur. I call these Cloud Native Applications. They have certain traits that should look a little familiar to cloud folks.

You cannot create a cloud native application doing things the same way you always have before. It simply will not work. The necessary software architecture and systems architecture has changed if you want your application to run on the cloud w/ no SPOFs.

Just needed to get that off my chest. Some related links for good reading:

http://blog.basho.com/2011/04/21/Amazons-outage-proves-riaks-vision/

http://www.thestoragearchitect.com/2011/04/22/so-your-aws-based-application-is-down-dont-blame-amazon/

http://highscalability.com/blog/2011/4/22/stuff-the-internet-says-on-scalability-for-april-22-2011.html

http://www.infoq.com/news/2011/04/amazon-ec2-outage

And if your REALLY keen to write some CNA’s (contact me) and read...

http://www.infoq.com/presentations/Actor-Thinking
http://www.infoq.com/presentations/1000-Year-old-Design-Patterns

]]>Cloud Native Applicationshttp://www.productionscale.com/home/2011/4/17/cloud-native-applications.htmlKent Langley2011-04-18T05:46:04Zcloud computing cloud computing distributed distributedThe Five Pillars of Monitoring – Pillar Onehttp://www.productionscale.com/home/2011/2/18/the-five-pillars-of-monitoring-pillar-one.htmlKent Langley2011-02-18T18:13:53Zbusiness monitoring