One of the things that people might not know about me, is my obsession to make my code as efficient as possible. Many people might not realize how much of a task or undertaking that this might be, but it is surely a task as monumental as climbing Mount Everest, except this time it is a challenge for the mind… In trying to make code efficient, there are many different factors that play a part – size of project or solution, tiers, language used, experience and training of the programmer, technologies used, maintainability of the code – the list can go on for quite some time.
I spend quite a bit of time when developing trying to determine what is the best way to implement a feature to accomplish the efficiency that I look to achieve. One program that I have recently come to learn about – Red Gate ANTS Performance (CLR) and Memory profiler gives me tools to accomplish that job more efficiently as well. In this review, I am going to cover some of the features of the ANTS memory profiler set by compiling some hideous example code to test against.
As a member of the Geeks With Blogs Influencers program, one of the perks is the ability to review products, in exchange for a free license to the program. I have not let this affect my opinions of the product in any way, and Red Gate nor Geeks With Blogs has tried to influence my opinion regarding this product in any way.
Introduction – Part 2
In my last post, I reviewed the feature packed Red Gate ANTS Performance Profiler. Separate from the Red Gate Performance Profiler is the Red Gate ANTS Memory Profiler – a simple, easy to use utility for checking how your application is handling memory management… A tool that I wish I had had many times in the past. This post will be focusing on the ANTS Memory Profiler and its tool set.
The memory profiler has a large assortment of features just like the Performance Profiler, with the new session looking nearly exactly alike:
ANTS Memory Profiler
Memory profiling is not something that I have to do very often… In the past, the few cases I’ve had to find a memory leak in an application I have usually just had to trace the code of the operations being performed to look for oddities… Sadly, I have come across more undisposed/non-using’ed IDisposable objects, usually from ADO.Net than I would like to ever see. Support is not fun, however using ANTS Memory Profiler makes this task easier. For this round of testing, I am going to use the same code from my previous example, using the WPF application.
This time, I will choose the ‘Profile Memory’ option from the ANTS menu in Visual Studio, which launches the solution in its currently configured state/start-up project, and then launches the ANTS Memory Profiler to help. It prepopulates all of the fields with the current project information, and all I have to do is select the ‘Start Profiling’ option.
When the window comes up, it is actually quite barren, just giving ideas on how to work the profiler. You start by getting to the point in your application that you want to profile, and then taking a ‘Memory Snapshot’. This performs a full garbage collection, and snapshots the managed heap. Using the same WPF app as before, I will go ahead and take a snapshot now.
As you can see, ANTS is already giving me lots of information regarding the snapshot, however this is just a snapshot. The whole point of the profiler is to perform an action, usually one where a memory problem is being noticed, and then take another snapshot and perform a diff between them to see what has changed. I am going to go ahead and generate 5000 primes, and then take another snapshot:
As you can see, ANTS is already giving me a lot of new information about this snapshot compared to the last. Information such as difference in memory usage, fragmentation, class usage, etc… If you take more snapshots, you can use the dropdown at the top to set your actual comparison snapshots.
If you beneath the timeline, you will see a breadcrumb trail showing how best to approach profiling memory using ANTS. When you first do the comparison, you start on the Summary screen. You can either use the charts at the bottom, or switch to the class list screen to get to the next step. Here is the class list screen:
As you can see, it lists information about all of the instances between the snapshots, as well as at the bottom giving you a way to filter by telling ANTS what your problem is. I am going to go ahead and select the Int16 to look at the Instance Categorizer
Using the instance categorizer, you can travel backwards to see where all of the instances are coming from. It may be hard to see in this image, but hopefully the lightbox (click on it) will help:
I can see that all of these instances are rooted to the application through the UI TextBlock control. This image will probably be even harder to see, however using the ‘Instance Retention Graph’, you can trace an objects memory inheritance up the chain to see its roots as well.
This is a simple example, as this is simply a known element. Usually you would be profiling an actual problem, and comparing those differences. I know in the past, I have spotted a problem where a new context was created per page load, and it was rooted into the application through an event. As the application began to grow, performance and reliability problems started to emerge. A tool like this would have been a great way to identify the problem quickly.
Overall, I think that the Red Gate ANTS Memory Profiler is a great utility for debugging those pesky leaks.
3 Biggest Pros:
- Easy to use interface with lots of options for configuring profiling session
- Intuitive and helpful interface for drilling down from summary, to instance, to root graphs
- ANTS provides an API for controlling the profiler. Not many options, but still helpful.
2 Biggest Cons:
- Inability to automatically snapshot the memory by interval
- Lack of complete integration with Visual Studio via an extension panel
Ease of Use (9/10) – I really do believe that they have brought simplicity to the once difficult task of memory profiling. I especially liked how it stepped you further into the drilldown by directing you towards the best options.
Effectiveness (10/10) – I believe that the profiler does EXACTLY what it purports to do.
Features (7/10) – A really great set of features all around in the application, however, I would like to see some ability for automatically triggering snapshots based on intervals or framework level items such as events.
Customer Service (10/10) – My entire experience with Red Gate personnel has been nothing but good. their people are friendly, helpful, and happy!
UI / UX (9/10) – The interface is very easy to get around, and all of the options are easy to find. With a little bit of poking around, you’ll be optimizing Hello World in no time flat!
Overall (9/10) – Overall, I am happy with the Memory Profiler and its features, as well as with the service I received when working with the Red Gate personnel.
Thank you for reading up to here, or skipping ahead – I told you it would be shorter! Please, if you do try the product, drop me a message and let me know what you think! I would love to hear any opinions you may have on the product.
Feel free to download the code I used above – download via DropBox
| posted on Monday, November 19, 2012 3:39 AM