Skip to main content

Premature optimization

I have been assigned the task to increase the the performance of the application in my last 2 companies and from my past experience I believe in 80-20 code execution rule, i.e. in production most of the time its the 20% of the code that is being executed and the rest 80% is used rarely. Therefore before you embark on increasing the performance of the website
  1. You should know your customers
  2. You should know their usage patterns
  3. You should first collect statistics about what are the features that customers are using heavily
Don't try to guess which features are used, do a quantitative analysis and focus on only those features that are used heavily. Apache/Tomcat access logs can be your buddies in reaching the goal, take daily logs from last 2-3 months put them in mysql or any DB using some simple python/perl script and create some excel graphs and do a trend analysis. For example in the below graph of webdav usage, I would rather focus on optimizing the PROPFINDs first then other operations.

Writing simple code is hard, following KISS (http://en.wikipedia.org/wiki/KISS_principle) is tough, whenever you try to increase performance of a feature you compromise simplicity and make the code a bit more complex then it was before. Therefore I don't believe in premature optimization, Donald Knuth had said "We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil".

Every programmer should follow some basic programming guidelines for performance when writing code but he shouldn't be too obsessive about performance when writing code, I have seen people writing Java code like C. They would not use Java's Date datatype or ArrayList and will use long to store dates and arrays instead of arraylist everywhere. Now tell me this if you are developing some Admin screens,user profile screens for a website or you are developing some internal application to be used by a handful of people making 100-200 request in an hour, why would you spend so much time making the code worse and complex by optimizing it prematurely. Analyze your apache logs insanely and focus on features that are used by customers heavily. Take the first 10 culprits optimize them and then repeat the process till you feel no more juice can be squeezed from the lemon.To optimize the top culprits if you have to compromise simplicity then its ok but don't make the job for the guy who is going to maintain the code after you leave hard by prematurely optimizing the whole code base.

Ofcourse if you are designing a new feature in an application and you foresee that it's going to be used heavily then  keep scalability and performance in mind when architecting the application but keep the code simple and before releasing it do some load testing using Jmeter or Load Runner and profile the code using some tool like JProfiler and optimize only the hotspots (code that is being executed 80% of the time ;) ).
Labels:

Comments

Post a Comment

Popular posts from this blog

Haproxy and tomcat JSESSIONID

One of the biggest problems I have been trying to solve at our startup is to put our tomcat nodes in HA mode. Right now if a customer comes, he lands on to a node and remains there forever. This has two major issues: 1) We have to overprovision each node with ability to handle worse case capacity. 2) If two or three high profile customers lands on to same node then we need to move them manually. 3) We need to cut over new nodes and we already have over 100+ nodes.  Its a pain managing these nodes and I waste lot of my time in chasing node specific issues. I loath when I know I have to chase this env issue. I really hate human intervention as if it were up to me I would just automate thing and just enjoy the fruits of automation and spend quality time on major issues rather than mundane task,call me lazy but thats a good quality. So Finally now I am at a stage where I can put nodes behing HAProxy in QA env. today we were testing the HA config and first problem I immediat...
Post a Comment
Read more

Adding Jitter to cache layer

Thundering herd is an issue common to webapp that rely on heavy caching where if lots of items expire at the same time due to a server restart or temporal event, then suddenly lots of calls will go to database at same time. This can even bring down the database in extreme cases. I wont go into much detail but the app need to do two things solve this issue. 1) Add consistent hashing to cache layer : This way when a memcache server is added/removed from the pool, entire cache is not invalidated.  We use memcahe from both python and Java layer and I still have to find a consistent caching solution that is portable across both languages. hash_ring and spymemcached both use different points for server so need to read/test more. 2) Add a jitter to cache or randomise the expiry time: We expire long term cache  records every 8 hours after that key was added and short term cache expiry is 2 hours. As our customers usually comes to work in morning and access the cloud file server it ...
Post a Comment
Read more

Spring 3.2 quartz 2.1 Jobs added with no trigger must be durable.

I am trying to enable HA on nodes and in that process I found that in a two test node setup a job that has a frequency of 10 sec was running into deadlock. So I tried upgrading from Quartz 1.8 to 2.1 by following the migration guide but I ran into an exception that says "Jobs added with no trigger must be durable.". After looking into spring and Quartz code I figured out that now Quartz is more strict and earlier the scheduler.addJob had a replace parameter which if passed to true would skip the durable check, in latest quartz this is fixed but spring hasnt caught up to this. So what do you do, well I jsut inherited the factory and set durability to true and use that public class DurableJobDetailFactoryBean extends JobDetailFactoryBean {     public DurableJobDetailFactoryBean() {         setDurability(true);     } } and used this instead of JobDetailFactoryBean in the spring bean definition     <bean i...
7 comments
Read more