 Okay. So we'll take a look now at the tracker updates for 237. What I wanted to emphasize at the very beginning of this is that for the individual data side of DHS2, which is tracker and events, we really have been focused very much on performance and stability for the last year and a half or so, which has been very timely because we've seen really large implementations coming along for first all of the COVID surveillance needs and then now really getting going with the COVID vaccine programs that are rolling out nationally. So the biggest changes that you'll notice in tracker for 237 are actually things we've also been rolling out. We've been rolling out into the patch releases as well and getting them back ported as far as we can throughout this time period. So there have been really significant changes to the tracker performance across the board for 235, 236, 237. I would really stress that if you are not on one of those versions of tracker and you're planning to use tracker for any large scale national implementations like the upcoming COVID vaccine trials, then you really do want to upgrade your instance to one of the newer versions. In fact, we're continuing to release performance improvements in the upcoming patch releases for 35, 36 and 37 that are going to be happening throughout December and January. One of the things that you'll see the biggest change in for 237 is how the ID schema are working for generating unique IDs. And we really would strongly recommend if you're doing a large scale national implementation that you want to be using a sequentially based ID schema rather than a random. But we can see dramatic improvements in how it impacts the performance of your tracker implementation. I was going to give you just some numbers here to give you a sense of what we're talking about in terms of performance. These are real numbers coming out of a national COVID vaccine campaign over the last month. We've seen that they've registered about 2 million people this month. And we're using screenshots here from Glow Roots, the open source monitoring application software that you actually could could apply to your own systems will be really useful for you to get a sense of how things are going as you start to scale up and use tracker on these on these large implementations. We have been able to get the system handling the entire load of this national COVID vaccine campaign, peaking at around 25,000 requests per minute. You can see here the large hump here during working hours for this COVID vaccine campaign. The search time that you have had some somewhere around 2.7,000 requests per minute in the search TI posting time around 230 requests per minute, the event data values. It's been quite significant how much the system is being used and how much it can handle with the changes that we've introduced into the software. I'm going to show you that with all of those transactions coming through about 99% of them take less than 66 milliseconds at this point with these performance improvements. And most of them are taking less even than 40 milliseconds. So this is a really great set of numbers to give you confidence that tracker can handle these very large implementations. But it is something that, again, you'll want to make the most of how you configure and set up your DHS to tracker so that it can perform to these kinds of levels. And again, you'll want to stay on as current of a release as you can for the version that you're using no older than 235. During this time period, we've also continued to create new documentation guidance and advice for you about how to set up these large scale tracker implementations. There will be more of this coming out to the community, but I wanted to take the opportunity with so many of you watching to mention some of the things that you can do to optimize your tracker. I already mentioned the 235 and up also minimizing the use of program indicators in your dashboard. I know that many, many systems have set up hundreds and hundreds of program indicators that they allow to just calculate each time a dashboard is open. That gives you a really heavy burden on your server when there are options for you to be able to push those aggregated numbers into the aggregate model instead. There's a new documentation for you on how best to do that. You also want to limit access to those dashboards that need to continue using program indicators, particularly if it's the default landing page for your system. Again, it creates a bunch of unnecessary stress on the server if you're loading these numbers over and over again for every person that's logging in. One of the things that can be done to optimize your performance again will be putting out more and more documentation about this. And the way that we have come to these improved numbers and the ability to handle this has been by working closely with some of the early countries that have been able to use the tracker for their COVID vaccine campaigns will continue to do so and make changes as we go. So again, your model as you're applying and rolling out national systems is really to stay current to do the patch updates and to make sure that you're staying on top of any of the implementation guidance that are coming out. So just showing you here some of the documentation that there is teaching everybody how to save your aggregated tracker data as aggregate data values, much more documentation available here in docs.dhs2.org where you can change up the way that you are running your analytics and make sure that everything is optimized to be able to handle what you need to be able to do in your implementation.