SYS-CON MEDIA Authors: Pat Romanski, Gary Arora, Zakia Bouachraoui, Yeshim Deniz, Liz McMillan

Blog Feed Post

How SproutCore Makes Your App Run Faster

If SproutCore 1.0 had a theme it would be “performance”. We’ve spent a lot of time - almost a year in fact - trying to make SproutCore applications the fastest web apps on the planet. I think we’ve done a pretty good job overall.

For example, we took Steve Souders excellent book on High Performance Websites and built his 14 rules right into our platform. We also discovered a few other rules along the way and built those in as well.

That means simply by adopting SproutCore for your web apps [and following our deployment recommendations], you will get best performance practices out of the box. No black-magic required.

So what did we do specifically to make SproutCore apps faster? Lots of things but here’s how we stack up against Steve’s recommendations:

Rule 1: Make Fewer HTTP Requests.

When you build your app in SproutCore, it takes all of your JavaScript and CSS files and generates “packed” files that include all your code in one file. Most SproutCore apps only need to load two stylesheets and two JS files when you go to production. Although its not automatic, the built-in SproutCore theme (Ace) is also sprited, keeping the number of image files down to 5-6.

Rule 2: Use a Content Delivery Network.

OK, we can’t bundle a CDN with our tools [yet], but all SproutCore resources have versioning built into the URLs. To add a CDN like Amazon’s CloudFront, just turn it on. Your entire SproutCore app will literally live on the CDN nodes, near your users instead of in your data center. Major perf gain. I can’t understate the importance of using a CDN to speed up your app.

Rule 3: Add an Expires Header.

Again, we can’t configure your web server for you, but when you build your SproutCore app, all of your assets will be packaged under a single directory with versioning. All you need to do is add the Expires header to your Apache config and your entire SC app will be cached in the users browser. Be careful though! Only set your expiration to a max of 1-year. Any longer and some proxies will actually ignore your header and turn off caching entirely.

Rule 4: Gzip Components

This is documented in our Deployment Docs. SproutCore’s JavaScript is already combined for you; but it is intended for you to Gzip it as well. Turning on Gzipping in your web browser can shave up to 500msec from your load time.

Rule 5: Put Stylesheets at the Top

Done. SproutCore generates an index.html that does this for you automatically.

Rule 6: Put Scripts at the Bottom

Also done. The index.html loads all your scripts last.

Although you can add scripts so they load at the top of your index.html if you jump through a few hoops, we have intentionally made designed the system to put scripts at the bottom whenever possible.

Rule 7: Avoid CSS Expressions

Actually, this is a more special cased version of a general rule that we use which is:

Don’t use complicated CSS layout.

Even though CSS will let you specify a rule to style the 3rd child of a 2nd cousin’s aunt’s grandmother, that doesn’t mean you should do it. Browsers have to invalidate and repaint larger portions of your screen to handle these kinds of crazy rules.

SproutCore is designed to encourage you to use simple absolute positioning for most app-like views. You can use regular “flow” layouts, but it should be restricted to actual document-like content such as help files and messages.

Absolute positioning is actually a lot more flexible than most people realize; it can even resize automatically with the browser. It is just a simpler, more easily predicted model for layout that is relatively inexpensive for the browser as long as you use it everywhere. It also encourages you to keep your CSS rules simple, which makes the browser faster.

And for the love of all things that are good, don’t use CSS expressions - the IE-only way to embed JavaScript into CSS. It is not only proprietary; it is also terrible for performance.

Rule 8: Make JavaScript and CSS External

Done. All SproutCore JS and CSS is saved as external files, ready for caching and CDN hosting as I mentioned earlier.

Rule 9: Reduce DNS Lookups

So, finally a rule where SproutCore really can’t really help you. For the record, the A records, not CNAMES, whenever possible.

Rule 10: Minify JavaScript

Done.

In fact if you ever wonder why an sc-build is taking so long to finish, it is probably the YUI minifier working on all of your JS and CSS to pack it down. There are faster solutions out there, but the YUI minifier did the best overall job of packing down your code into the smallest size possible in our tests, so its the preferred choice for the moment.

Rule 11: Avoid Redirects

Done.

Since SproutCore apps are single-page applications, the user rarely will need to wait on a page to load, let alone a redirect. The one exception to this is when you need to handle login. Sometimes best practice will require you to redirect over to a domain under HTTPS to handle login. This should be a rare occurrence.

Rule 12: Remove Duplicate Scripts

Done.

SproutCore’s build tools will detect when your various frameworks depend on other frameworks and then build a single dependency chain so that each script is loaded only once and in the proper order.

It is possible to get around this of course, but you would have to work at it. That’s the point; make it easiest to do things in a performant way.

Rule 13: Configure ETags

We can’t do this in the SproutCore itself but the built files are ETag ready. Our deployment guide also includes this in the checklist of recommendations.

Rule 14: Make Ajax Cacheable

This rule mostly applies to your dynamic requests, but when you do load resources from SproutCore dynamically, they are setup to follow the standard caching rules so they can load quickly.

Other Rules

I said we came up with a few more rules of our own.  Most of these relate to building heavy client-side applications, but they matter nonetheless.  Here’s a short list of things that make SproutCore apps fast:

Rule 15: Don’t Do More Work Than Necessary

SproutCore apps do very little work before they get your app up and running.  For example, the entire view layer of your application may be loaded right away but no views will be constructed until you actually try to bring them onto screen.

We also added support for something called a RunLoop, which allows you to defer actions until just before you are finished running your JavaScript.  For example, if you make a view visible, then not visible, then visible again, SproutCore won’t actually update the DOM until just before the event handler returns.

By following this principle, your app will generally load and launch very quickly and will have more even performance over the long run.

Rule 16: Avoid Touching the DOM

Even though the DOM is available to you in JavaScript, underneath the covers it is actually implemented in straight C.  All browsers have some kind of bridge you need to cross every time you access the DOM to get from the JavaScript world to the C world.

This means you pay a performance cost for every DOM access; even for something as simple as reading the “id” from an element. (i.e. var foo = element.id).  In Internet Explorer this effect is especially pronounced since the COM model that wraps mshtml leaves 10 or more layers of code between the JS and the DOM.

To work around this, SproutCore’s entire view layer has been specially designed to minimize the number of times we touch the DOM at all.  The first time you generate your UI for example, SproutCore will generate one large HTML string and put it on screen using “innerHTML”.

Compare this to frameworks like Cappuccino, jQuery plugins that work mostly by constructing like DOM elements, or even SproutCore before 1.0, and you’ll see sometimes a 100x performance improvement.  It’s huge.

Of course, once you’ve generated your UI for the first time, SproutCore will need to work more directly with the DOM so we don’t get as much of an improvement all the time.  Overall, however, SproutCore 1.0 improved our view layer performance by 10x over the 0.9 version (which was pretty fast to begin with).

Rule 17: Don’t Load More Than Necessary

If you build a very large application, you will often end up with lots of UI that may not be needed right away - or possibly ever - during a typical session.  Even though SproutCore avoids doing more work than necessary on page load, you can still shorten your load time even further by not downloading the code for this UI at all.

SproutCore 1.0 sports a new “dynamic framework” model that allows you to lazily load frameworks only when they are needed instead of all up front.  Some apps have been built with dozens of such frameworks, cutting their load time by almost 30%.

And More…

There are more enhancements we’ve made here and there - such as tightly controlling memory usage and managing change propogation in the app.  There are more improvement we will make over time as well, but I think you get the idea.

The #1 reason I hear people say they can’t build client-side web apps is because they are worried about performance.

Their concerns are not totally unfounded.  It has traditionally been very hard to build large JavaScript applications that perform well.  The reason, though, is not because the browsers can’t handle it.  It’s just because most of us didn’t know all the black magic needed to do it right.

My hope is that SproutCore 1.0 will lay this concern to rest for good.  You still have to think about performance, but SproutCore takes care of the most important stuff for you.  Apps can have lots of features and still perform very well, thank you very much.

There are a lot of cool features we could have focused on for 1.0 - more controls for example or a nice visual UI builder.  We are actually working on these [especially the view builder], but none of it really matters of your app performance sucks.  You have to get that right first.

Great web performance.  Built-in.  Now that’s a feature.

Read the original blog entry...

More Stories By Charles Jolley

Charles Jolley is the creator of SproutCore and CEO of Sproutit, which provides consulting and software services for building SproutCore applications on the web. He has helped build some of the most popular RIAs on the web today.

Latest Stories
While a hybrid cloud can ease that transition, designing and deploy that hybrid cloud still offers challenges for organizations concerned about lack of available cloud skillsets within their organization. Managed service providers offer a unique opportunity to fill those gaps and get organizations of all sizes on a hybrid cloud that meets their comfort level, while delivering enhanced benefits for cost, efficiency, agility, mobility, and elasticity.
Isomorphic Software is the global leader in high-end, web-based business applications. We develop, market, and support the SmartClient & Smart GWT HTML5/Ajax platform, combining the productivity and performance of traditional desktop software with the simplicity and reach of the open web. With staff in 10 timezones, Isomorphic provides a global network of services related to our technology, with offerings ranging from turnkey application development to SLA-backed enterprise support. Leadin...
DevOps has long focused on reinventing the SDLC (e.g. with CI/CD, ARA, pipeline automation etc.), while reinvention of IT Ops has lagged. However, new approaches like Site Reliability Engineering, Observability, Containerization, Operations Analytics, and ML/AI are driving a resurgence of IT Ops. In this session our expert panel will focus on how these new ideas are [putting the Ops back in DevOps orbringing modern IT Ops to DevOps].
Darktrace is the world's leading AI company for cyber security. Created by mathematicians from the University of Cambridge, Darktrace's Enterprise Immune System is the first non-consumer application of machine learning to work at scale, across all network types, from physical, virtualized, and cloud, through to IoT and industrial control systems. Installed as a self-configuring cyber defense platform, Darktrace continuously learns what is ‘normal' for all devices and users, updating its understa...
Enterprises are striving to become digital businesses for differentiated innovation and customer-centricity. Traditionally, they focused on digitizing processes and paper workflow. To be a disruptor and compete against new players, they need to gain insight into business data and innovate at scale. Cloud and cognitive technologies can help them leverage hidden data in SAP/ERP systems to fuel their businesses to accelerate digital transformation success.
Concerns about security, downtime and latency, budgets, and general unfamiliarity with cloud technologies continue to create hesitation for many organizations that truly need to be developing a cloud strategy. Hybrid cloud solutions are helping to elevate those concerns by enabling the combination or orchestration of two or more platforms, including on-premise infrastructure, private clouds and/or third-party, public cloud services. This gives organizations more comfort to begin their digital tr...
Most organizations are awash today in data and IT systems, yet they're still struggling mightily to use these invaluable assets to meet the rising demand for new digital solutions and customer experiences that drive innovation and growth. What's lacking are potent and effective ways to rapidly combine together on-premises IT and the numerous commercial clouds that the average organization has in place today into effective new business solutions.
Keeping an application running at scale can be a daunting task. When do you need to add more capacity? Larger databases? Additional servers? These questions get harder as the complexity of your application grows. Microservice based architectures and cloud-based dynamic infrastructures are technologies that help you keep your application running with high availability, even during times of extreme scaling. But real cloud success, at scale, requires much more than a basic lift-and-shift migrati...
David Friend is the co-founder and CEO of Wasabi, the hot cloud storage company that delivers fast, low-cost, and reliable cloud storage. Prior to Wasabi, David co-founded Carbonite, one of the world's leading cloud backup companies. A successful tech entrepreneur for more than 30 years, David got his start at ARP Instruments, a manufacturer of synthesizers for rock bands, where he worked with leading musicians of the day like Stevie Wonder, Pete Townsend of The Who, and Led Zeppelin. David has ...
Darktrace is the world's leading AI company for cyber security. Created by mathematicians from the University of Cambridge, Darktrace's Enterprise Immune System is the first non-consumer application of machine learning to work at scale, across all network types, from physical, virtualized, and cloud, through to IoT and industrial control systems. Installed as a self-configuring cyber defense platform, Darktrace continuously learns what is ‘normal' for all devices and users, updating its understa...
Dion Hinchcliffe is an internationally recognized digital expert, bestselling book author, frequent keynote speaker, analyst, futurist, and transformation expert based in Washington, DC. He is currently Chief Strategy Officer at the industry-leading digital strategy and online community solutions firm, 7Summits.
Addteq is a leader in providing business solutions to Enterprise clients. Addteq has been in the business for more than 10 years. Through the use of DevOps automation, Addteq strives on creating innovative solutions to solve business processes. Clients depend on Addteq to modernize the software delivery process by providing Atlassian solutions, create custom add-ons, conduct training, offer hosting, perform DevOps services, and provide overall support services.
Contino is a global technical consultancy that helps highly-regulated enterprises transform faster, modernizing their way of working through DevOps and cloud computing. They focus on building capability and assisting our clients to in-source strategic technology capability so they get to market quickly and build their own innovation engine.
When applications are hosted on servers, they produce immense quantities of logging data. Quality engineers should verify that apps are producing log data that is existent, correct, consumable, and complete. Otherwise, apps in production are not easily monitored, have issues that are difficult to detect, and cannot be corrected quickly. Tom Chavez presents the four steps that quality engineers should include in every test plan for apps that produce log output or other machine data. Learn the ste...
Digital Transformation is much more than a buzzword. The radical shift to digital mechanisms for almost every process is evident across all industries and verticals. This is often especially true in financial services, where the legacy environment is many times unable to keep up with the rapidly shifting demands of the consumer. The constant pressure to provide complete, omnichannel delivery of customer-facing solutions to meet both regulatory and customer demands is putting enormous pressure on...