How Do You Say "Ehcache?"

What's the most pressing issues at Terracotta? A challenge that has been dogging us for years? Why of course it's teaching people the proper pronunciation of "Ehcache." Have no fear. Here is a video that will help you to get it right!

Click here and be educated!

What's Apache licensed, improves performance, reduces tuning and is now GA?

I'm glad you asked... Ehcache 2.5 with ARC went GA today. At a couple of megs Ehcache with ARC is lightweight and helps with real tuning and performance issues in Java applications.

Like Ehcache itself I'll keep this blog lightweight.

Here is why you care:

• How Much/No OOME's - Specify how much data you want to cache in bytes for on heap unserialized objects. Ehcache keeps the cache under that size to avoid OOME's
• Runtime, Tune as needed - Specify total heap to use for cache and then, when you feel like it, tweak individual caches for optimal performance. 
• CacheManager - Max Bytes on heap (Percentage of heap or fixed number)
• Cache - max bytes on heap (Percentage of CacheManager or fixed number)
• Lots of caches made easy - Use hundreds of caches without sizing each one individually
• Hibernate people, this means YOU.
• Did I mention NO OOME's
• Fence your heap - Like Virtualized Machines give your app's internals a percentage of the total heap and your users a percentage. Prevent the two from stomping each other.
• People who ship software to others TAKE NOTE.
• Pinning - Pin caches or entries to heap that need to always be available and extremely fast.

Who Should Care?

• Hibernate users
• Lots of caches, impossible to tune each individually
• Spring Annotations users
• Pick the amount of heap you want to dedicate to caching and then just mark the methods that need caching
• Software Shippers (Frameworks, Servers or Apps)
• Don't force users to size caches
• Reserve part of the heap for your internal state and assign a safe size for users
• Application Admins
• Monitor an application in production and do sizing/pinning where needed on the fly

Conclusions

Ehache and ARC are free, lightweight and likely already in your application (embedded in Spring, Hibernate, Cold Fusion and countless other servers and frameworks.) This is the most requested feature set in the history of Ehcache. 

Give it a try!

More Info:

http://www.terracotta.org/products/enterprise-ehcache/automatic-resource-control

http://ehcache.org/documentation/configuration/cache-size

http://scaleaholic.blogspot.com/2011/09/spring-performance-with-annotations-and.html

http://scaleaholic.blogspot.com/2011/07/easy-java-performance-tuning-manage.html

http://www.dzone.com/links/r/new_efficient_ehcache_automatic_resource_control.html

Spring Performance With Annotations and Ehcache ARC

The goal of this blog is to highlight Ehcache's new ARC feature using Spring Annotations.

Terracotta is releasing Ehcache 2.5. It's biggest new feature is ARC (Automatic Resource Control). ARC brings runtime performance tuning and caching to the Systems Administrator while simplifying it for the Developer. I've written a couple of blogs describing what it is and the value it brings:

• Local Caching++ 
• Easy Java Performance Tuning: Manage Objects By Size On The Java Runtime Heap

These two blogs and the Ehcache docs are excellent sources of information on these highly requested/useful new capabilities. 

Getting Started

I wanted to give a brief demonstration of how someone would leverage Ehcache ARC in practice so I pinged the guys who do Spring Annotations and they OK'd me to leverage one of their samples. Spring Annotations enable one to mark methods that do performance sensitive operations to be cached to improve the application's speed. 

In order to follow along with me you'll want to read up on and check out the Spring Annotations sample here: 

http://code.google.com/p/ehcache-spring-annotations/wiki/WeatherServiceExample

Download it and play with it a bit to get familiar with how it works.

This sample is fairly self contained. It show's how to use the various annotations to improve performance in a simple and concise way. What it doesn't talk about is the configuration of the underlying cache in use. This ends up being very challenging in practice. Without help you'll run into at least a few of these 5 problems:

• Crash - Out of Memory Errors can crash your Application
• Pause - Long GC's will pause your application leading to an unpredictable user experience
• Wasted Space - In order to avoid the above two problems you over provision memory wasting ton's of space
• Tuning Hell - Either you or your users spend ton's of time making arbitrary decisions trying to balance out the memory usage of your caches to avoid the above 3 issues
• Poor Performance - Caches don't really help because they are incorrectly sized

Let's look at the ehcache.xml that ships with this sample:

The important parts of this config are from lines 30-31. It sets maxElementsInMemory to 100 and timeToLiveInSeconds to 300.

Why do these need to be set? Let's look at them individually:

maxElementsInMemory=100

When you started your Tomcat instance some amount of memory was reserved in the command line (Xms Xmx). If you didn't set a maxElementsInMemory your cache would grow infinitely until long GC's and then an OOME occur. So maxElementsInMemory is a resource management choice. But why 100? Is it because you know how big your entries are and you know 100 is exactly some percentage of the heap? More likely it's a guess. Probably a low ball guess so that you don't OOME wasting a whole bunch of heap. Now say you have 10's or 100's of caches that hold objects of varying size. How much harder is it to set the maxElementsInMemory now?

timeToLiveInSeconds = 300

This is a data freshness control. If your using an unclustered cache and multiple nodes then objects are likely being updated in a DB and the data in your cache can get out of date. How out of date? Depends what you set your timeToLiveInSeconds to.
NOTE: Keeping all nodes up to date is one of the reasons people use clustered caches.

Configuring a cache using the above controls makes an application sensitive to use case changes, changes in JVM heap settings, and number of caches. It also leaves the user of the application with a choice of leaving a lot of headroom to handle the worst case, risking OOME's and forcing the admin or user to tune at every deployment.

Ehcache ARC Gets Performance Without All The Tuning

These are the problems that Ehcache ARC, the most requested feature in Ehcache history, are designed to eliminate. Ehcache ARC makes your cache or caches controllable using the same metric the JVM's are started with. The same resource you are trying to manage. Bytes! In addition, instead of having to tune each individual cache you can instead make a high level choice at the CacheManager level and let ARC balance the resources for you.

In this change to the sample we have removed the cache entry count and instead allowed the cache to use 50% of the JVM heap (look at line 4 where it says maxBytesOnHeap). This setting tells the cache manager to allow for up to 50 percent of the heap for caching (Can also be configured using a fixed number of bytes instead of percentage). It will balance this heap across what ever caches are defined with this manager. This allow's an application to manage the available resources without waste and without GC pauses/OOMEs.

I updated and rebuilt my spring-annotations to use 2.5.0 and then rebuilt the sample using my newly created version of Ehcache Spring Annotations. You can grab the updated Jar here though I suspect the Spring Annotations guys will update the official build soon.

By using this approach as you add more and more @Cachable annotations to your application you don't have to change anything about the caches you create in ehcache.xml. They make use of the available cache resources as provided.

Wrapping Up


It has always been pretty easy to integrate a cache into an application. Things like Spring Annotations move the bar on that simplicity even further. The challenge was to make those caches effective. How does one get the most performance out of the resources available. Ehcache ARC was created to solve that problem for System Administrators, OEM's Developers and the entire OSS community. We are looking forward to people getting started with it and hope to receive lots of feedback.

What Is Terracotta?

One of the biggest challenges in software is telling people what your software is in a way that helps them make decisions about it. This challenge can often be as difficult as designing and building the software itself. The Terracotta team recently hooked up with the gang from Epipheo studios to create a 2 minute video to do just that. We spent a lot of time thinking about and refining things in order to get a clear succinct message. I think it came out quite well.

Check it out:

Easy Java Performance Tuning: Manage Objects By Size On The Java Runtime Heap

The Problem?

For those of you who use caching with frameworks like Hibernate, Spring or anything else, you know what a pain performance tuning can be. The tools you have available for tuning, resource management and avoiding OOME's boil down to counts and age controls. These are actually not resource management controls at all. They are data freshness controls and should be treated as such.

How do you even begin to figure out how many entries to allow in each Hibernate cache when you have a hundred of them? What if things change and objects get bigger, or smaller. What if you change your heap size and/or usage patterns?

You've Gotta Try This...

With Ehcache 2.5 you can just specify a percentage of heap or a heap size in bytes that your graphs of Java objects are allowed to use. This is accomplished by passing a simple size description into a Cache or CacheManager. All objects held onto by the cache will borrow their space from the cache's or manager's specified pool. Entries will get evicted from the cache as space runs low without any intervention from the developer. This can be done at the Cache Tier level whether it's on heap, disk or BigMemory. It's another way to reduce tuning, improve performance while avoiding OOME's. Learn more here.

If you do things in config or in code it's just a one line change:

A few More Details...

It works on any 1.5 or 1.6 JVM (Tested on Oracle, JRocket and IBM). Doesn't require any object serialization for the on heap management.

Learn More:

http://ehcache.org/documentation/configuration.html#Memory_Based_Cache_Sizing_Ehcache_2.5_and_higher

http://www.terracotta.org/

http://dsoguy.blogspot.com/2011/04/biggest-problems-in-local-caching.html

Try it out and give us lots of feedback on the Terracotta Forums:

http://forums.terracotta.org

Exciting Times... Terracotta and Software A.G.

Exciting times over here at Terracotta.

Read all about it:

http://www.softwareag.com/corporate/company/terracotta.asp

http://terracotta.org/terracotta-joins-software-ag

http://blog.terracottatech.com/2011/05/terracotta_joins_software_ag.html

This is great news for Ehcache, Quartz, and Terracotta customers and communities!

Local Caching++

Ok, so you've built your application in Java. You've used all the usual tools. Tomcat, Spring, Ehcache, Quartz, etc. Or maybe you went the JRuby, Grails or Scala route. You test your new application or hand it off to run in production and it's too slow. This is just a single node application at this point. It services 20-100 users. It's churning and burning the database, creating and recreating the same Web Pages, Users and other relevant data. You want to start caching locally to solve your latency and throughput problems. Then, upon a more detailed look at your application, you get scared.

You find your application has:

• 40 DB tables in Hibernate that can be cached
• A web cache
• A user session cache

Then it hits you. Caching is easy but cache tuning is hard.

What Makes Cache Tuning Hard?

In conversations with 100's of cache users there are actually a small handful of difficult to work through challenges applying caching to an application:

• Hibernate/Lots of caches - When using Hibernate you often end up with as many as 100 tables in your DB. How do you balance a fixed amount of resources(Heap/BigMemory) across 100 caches?
• Indirect knobs/Bytes vs Count/TTL - In local Java caching the control points are almost always measured in number of entries and time to live. But wait a minute! When I start the JVM I don't say how many objects the heap can hold and for how long. I say how many bytes of memory the heap can use?
• Who Tunes and When? - At some companies the desire is to have the "Application Administrator" do the tuning. At others it's the "Developer." They have different understandings of the application. The developer can tune by knowledge of the application. The app admin can only tune based on what's happening when the application is running.

These are the challenges we are working to solve in the next version of Ehcache. While it's early days on the dev side we would love feedback on our approaches. You can get a sense of how it's going to work from the doc on ehcache.org.

What We Are Building

Greg, me and the dev team spent a bunch of time pondering the above problems over the last few years. We felt that with two key improvements to how people tune we could address most of the above (and a few more items hit the rest):

• Tune from the top - Define max resource usage for the whole cache manager and then optionally define it for the individual caches underneath it as needed. So if you have a hundred caches you can start with, "Give these 100 caches N amounts of Heap/OffHeap." Then monitor and see if any specific caches need special attention.
• Tune the constrained resource, Bytes - TTL is a cache freshness concern not a resource management concern. Max entry count does not directly map to available heap resources. So we are adding "bytes" based tuning. This eliminates the mistake prone process of trying to control resources by TTL/TTI/count and hope you get it right. Instead you say, "I want to allow caching to use 30 percent of heap." We take it from there.

Wrapping Up

With those two key improvements a developer or admin is now directly turning the knob (Size of cache in bytes) that maps to the resources available in the JVM and doing it at a global level or a local level as needed to avoid hard to tune individual cache constraints.

This will work with all of our JVM level cache tiers (onHeap, BigMemory, Disk).

When you put those features together with other items coming in the next major release like entry and cache pinning and a snapshotting bootstrapper for warming we feel like this will be a very powerful release.

Please check out the new docs and give us feedback by commenting on this blog or posting the the Ehcache forums.

Help us make Ehcache as easy to use and powerful as we possibly can.

http://ehcache.org/documentation/configuration.html#Memory_Based_Cache_Sizing_Ehcache_2.5_and_higher