What is LMS server latency, and how it affects your elearning?

Posted by Greten on 25 May 2020 under Terms, Tips

The term server latency refers to the amount of time it takes for a signal to travel from the user's device to the server and vice versa. Server latency is an essential consideration in developing websites and web-based applications. Typically expressed in milliseconds, it depends on several factors such as the number of files requested by the browser or app, the number of users accessing the website or app at once, and the physical distance between the user and the server.

An LMS server latency is the amount of time it takes to communicate between the elearning module and the server where the learning management system (LMS) is stored. The elearning module corresponds to the server for several different requests, such as a request for an image or audio file, send an exam scores so the LMS can keep a record, and transmit information about every slide accessed and interactions.

LMS server latency within the elearning operations

LMS server latency icon: a laptop, a server, and two arrows circling on their back. There's also a clock in the middle of the circling arrows.The lower is the LMS server latency, the better it is for elearning experience.

Allow me to share an experience. Before, I do not even know this LMS server latency. Since you upload elearning modules as zip files, my impression was that the LMS would also send the elearning to the learner's computer as a zip file and then unzip it there.

As it turns out, I was incorrect. The elearning module calls the files as needed. If you're viewing slide 3 and an image file is needed in slide 4, then the module will request that file only when you're already viewing slide 4.

Disasters can be opportunities at times. I would not have known how the LMS server and the learner's device interact if the Wuhan virus (officially called COVID 19) never reached a pandemic level, necessitating many businesses, including my current employer, to switch to work-from-home. This situation, combined with my home DSL connection suddenly failing a week before the lockdown, forces me to do my work using less powerful mobile data. Before the pandemic, I was testing elearning modules through the company's powerful internet connection. Then, in all sudden, I have to test elearning modules using mobile data.

The result is uncanny. I can see the audio playing on a blank slide or slides with incomplete graphics. Sometimes, I can see animations and closed captions but no audio.

I searched the internet on what's going on with the learning modules I am testing and how to possibly remedy them, and it was through my search that I encountered the term "LMS server latency".

What causes a longer LMS server latency?

The causes of a longer LMS server latency is the same as that of website and web applications: too many requests or signals being sent to and receive from the LMS server. However, the reason the learning module or LMS server needs to send such requests can be different.

Too many files

The way server latency works in webpages is like this: if I have a choice between using a 90-kilobyte image and a set of three 30-kilobyte images, the 90-kilobyte image is better for server latency because the browser needs to send only one request to the server. The CSS sprites that web developers use loads faster because the browser sends only one request and receives only one image that is going to be used as three or more icons.

Using a Flash SWF instead of HTML5 can keep a lower the LMS server latency. An SWF animation and interactive can be only one file, while the HTML5 that does the same is a collection of images, audio files, and javascript files that make them work together. However, since Flash is almost no longer supported due to its vulnerabilities to malware, elearning developers have to keep using HTML5 while keeping the requests to the server at a minimum.

Too many interactions to send to LMS

File requests are not the only signals that the learning module sends to the LMS server. It may send data about the learner's every interaction with the module, what slide they are currently seeing, and their quiz score.

The LMS having a record of each action of the learner can be crucial to certain functions of the elearning, such as keeping the percentage already finished in case the learner's device suddenly shuts down, the browser suddenly closed, or some other unexpected bug. That way, the learner can know where to restart once he or she logged-in to the LMS again, or the LMS can just redirect the learner there.

Too many learners accessing the LMS at the same time

If too many learners using the LMS at once, the result can be longer LMS server latency; thus, an elearning that is not working as expected. In an asynchronous elearning, this is not usually expected because there's no single time that all or a large percentage of learners are accessing the LMS at the same time.

Still, there can be situations where too many learners may end up using the LMS simultaneously. For example, if most of your learners are the typical employees working nine-to-five jobs, many of them are using the LMS at around dinner time.

Large distance between the LMS server's location and the learner

Although not much of an issue due to faster internet connections nowadays, the distance between the learner and the LMS server can also contribute to higher LMS server latency.

Hence, it is generally good for learners to enroll in elearning courses run by a local elearning company. Elearning companies who want to have international scope should have redundant servers in various locations around the world.

The learner's internet connection

The learner's internet connection can also cause lag in delivering the elearning content. Some of the other causes of longer LMS server latency will not have much impact if not for the slower internet connection. You cannot notice the difference between a single 90-kilobyte file and a set of three 30-kilobyte files if your connection is at 35 Mbps (megabytes per second).

How to reduce LMS server latency?

To reduce the LMS server latency of your LMS and elearning modules, you need to address the causes of longer LMS server latency and work around them.

Reduce the number of files

Since Flash is no longer a viable option, you need to develop your elearning such that there are fewer files in the SCORM zip file as possible. Some of these solutions are as follows.

  • Does your elearning contain a continuous animation across several slides and no interactivity aside from the quiz at the end? Then, it's better to publish your elearning as a video. You can put the quiz in a separate elearning module that the learner can take later. Note, however, that not all elearning authoring tools can publish as video. Captivate 2019 can publish as video, but Storyline 3 cannot.
  • If you have a diagram, concept map, flow chart, or any figure that is made of several images, combine them into one large image using an image editor (e.g., GIMP, Photoshop) or vector graphics editor (e.g., Inkscape, Illustrator). Then, put them as one larger image file on the elearning slide.

Reduce the number of interaction to LMS

As mentioned earlier, you can reduce the number of interactions between the learning module and the LMS server by configuring the module to send only the quiz result at the end of the module. Put the quizzes at the end of the module or in separate quiz-only modules. Then, configure the module or modules, so they do not need to send to the LMS server the interactive actions the learners took or the slide number they are currently viewing.

There are options to report only the quiz results instead of the progress in slide number for both Adobe Captivate 10 and 11 (also called Captivate 2017 and 2019 respectively) and Articulate Storylines 2 and 3, but I will no longer cover the steps in this post. Perhaps, in some of my later posts. For now, you can search these easily; search "Storyline LMS reporting" or "Captivate Send Tracking Data At End" (without quotes) depending on which elearning authoring tool you are using.

It would help if you also weighed the benefits of reduced LMS server latency vs. the need for tracking. If your course is made of several short elearning lessons, then you do not need to track your learners' progress in terms of slide number; they can always go back to the previous lesson or go to the next lesson as needed. However, if the lesson lasts longer, say more than 10 minutes, you might need to track the progress of your learners. Moreover, you might want to track your learners' behavior towards the interactive elements in your lessons if you have a system that's doing data crunching to determine the best way to reinforce learning and improve the lessons.

Reduce the number of learners accessing the LMS at the same time

The overload due to simultaneous usage is a rare issue to asynchronous elearning because the learners decide what time they will take the lessons. However, should you encounter this issue, below are some of the possible solutions.

  • Schedule when the learners can use the LMS. In the earlier example of employees working a nine-to-five job, you can probably schedule some of them at 6:00PM to 7:00PM time slot, and others at 7:00PM to 8:00PM time slot.
  • Although this should be the last resort or implemented only if the LMS server is old, you should upgrade your LMS server if there's a frequent lag due to several learners using the LMS at the same time.

Working around the LMS server latency due to distance from the server

If the distance between the server and the learner is so far that it causes a huge lag in elearning, consider the following options.

  • If your elearning company receives learners from different parts of the world, consider setting-up redundant servers in several locations, preferably within or near the countries where many of your learners are based.
  • If the elearning was developed only for the use of one company, consider putting it in the local network or intranet instead of the internet.
  • For learners, they might want to consider enrolling only to local elearning providers. However, it does not follow that a local elearning company has a server closer to the learner. It's possible for an elearning company to do business in one country, but have their servers located in another country. If you're an owner or manager of an elearning company and it's not cost-effective to setup a local server, then look for a server provider than can transmit fast to your or your learner's location even if that server is outside your country.

Dealing with learner's slower internet connection

As an elearning developer or someone who works in an elearning company, you can only influence the first four causes of long LMS server latency: number of files, number of server requests, number of users accessing simultaneously, and distance from the server. As for the internet connection, only the learners can solve that. They can upgrade their internet connection or look for another provider with a reputation of providing faster internet.

However, if you did more to minimize the effect of the first four causes, then the probability of bad learning experience due to laggy elearning or LMS would be significantly reduced.

Conclusion

An LMS server latency is the time it takes to communicate between the elearning module and the server in which the learning management system (LMS) is stored. The larger is the LMS server latency, the more noticeable is the lag in elearning. You can reduce the LMS server latency by reducing the file objects in your elearning module, interaction with the LMS server, the number of learners using the LMS simultaneously, and the distance between the learners and the physical server.

Bibliography:

Last updated on 25 May 2020.

Share your thoughts

* Required. Your email will never be displayed in public.

Instructional design and educational technology for effective learning