Duration 25:02
16+
Play
Video

Optimally deploy an application cache with Memorystore

Gopal Ashok
Product Manager at Google Cloud Platform
  • Video
  • Table of contents
  • Video
Google Cloud Next 2020
July 14, 2020, Online, San Francisco, CA, USA
Google Cloud Next 2020
Request Q&A
Video
Optimally deploy an application cache with Memorystore
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Add to favorites
1.07 K
I like 0
I dislike 0
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
  • Description
  • Transcript
  • Discussion

About speaker

Gopal Ashok
Product Manager at Google Cloud Platform

Product management leader passionate about creating impactful products Track record of planning, delivering and marketing enterprise products Proven leadership skills with emphasis on empowering people and driving data-driven decisions Strong analytical skills with a record of developing short term and long term product strategy Fast learner with the proven ability to learn new technology areas from a technical andmarket perspectiveSpecialties: Product Management,Product Strategy, Program Management, Database, High Availability, Enterprise software; Messaging and Positioning,Customer relationship, Public speaking

View the profile

About the talk

Memorystore provides a managed in-memory store to speed up application performance. Learn about exciting new Memorystore features and how to optimally deploy it for various application use cases. See how customers are deploying Memorystore, the challenges they’ve faced, and best practices they used to run an optimal caching layer for their use cases.

Speaker: Gopal Ashok

Watch more:

Google Cloud Next ’20: OnAir → https://goo.gle/next2020

Subscribe to the GCP Channel → https://goo.gle/GCP

#GoogleCloudNext

DBS210

product: Cloud Memorystore for Redis, Cloud Memorystore for Memcached; fullname: Gopal Ashok;

event: Google Cloud Next 2020; re_ty: Publish;

Share

Hello, my name is gopal Ashok, I'm the product manager for cloud memory store. We going to talk about, optimally deploying an application cache with memory store. There are a few areas that I want to cover today from Danville to talk, a little bit about redis and why you should be using when you aren't applying a memory store for your application, you ski. So, I think we all know in memory stores are a critical part of modern application, architecture and processing. Data in memory, it is impossible

to determine user retention. So have a good memory store that can deliver the the speed that is required by modern applications is supercritical. The other aspect of it is that it also in a way or the other memories or comes into play along with speed in a database are back in store. For example mission on down side or right side of your back and still provisioning, this can help you manage the overall cost for data access. so as you can see, in memory store, hat is a super critical component to Enable your application to steal and provide the

speed and responsiveness at the application. Example of this is a very common example of using sections are using a second store using a back-end database, ask perception store. If you are using a back in store for back-end database, a secure Second Story for a certain amount of time, looking for a call so you can store your sessions and restore and basically the same time and the processing is done in memory. And processing happens within milliseconds, your overall

Supercenter in Forest today So let's take a look at one of her choices when it comes to the most popular, search a website for waiting, databases and Elsa, and the stack, Overflow survey uses the most locked database. Why is red is so popular. Running has been extremely fast in memory, stored up for whites, only second all sing, but more importantly provides service light on top of that is also for streaming, read aloud application, and abuse cases. At the very least, it is becoming the number one choice for any memory stores for building Mauldin

application. So if you're migrating gcbro of fully managed Rena Sandman casts on PC. So if you are you are you know? I'm the key thing about having a memory stored as a fully managed service is that it takes away the burden of managing Take a look at some of the key features and capabilities are memory store. The highest level. As I mentioned memory, supervisor fully matters, right inside Minecraft. And the other thing is that if you want to do it yourself,

then you have to maintain a proxy automatic security ensure that the inspector has deployed in a very secure environment. That is highly available and also highly secure, and have to spend time managing your focus on building your application. Let's take a closer look at some of the capabilities that before wine for memory, Sole for red cabbage. Here is a highly available in here, is that both of us are essentially, single primary so that means we don't support scale art architecture yet. So so

when you architect your application, that is something that you should keep in mind and I'll talk more about that. Bought from the store as a 30-goal. Our primary store you can take back from the ship's registration Park Supply memories or using a Chevy PC Network today, but now you'll be able to do it. I have to go to Jenny, Bush at BCC. We're excited to announce that we have. We are providing support for memcached Diaz Spotify. That is available with memories show.

Dating Apostles in and out. You always, you always know the very convenient service that allows you to show me a super excited about seeing them casting for you. You're currently in beta. Okay, so we talked about a few things we touched upon why Katherine is important. We touched upon some of the benefits of using. I want to talk about for the remaining Look at these cases is, especially to keep things on the sound is that each of these used for different items

that is expected from the Gathering layer of these patterns. For the application when it comes to and also be on site around. And maybe a small performance, is not significant from an application for Spectre in our Men cast are the basic fear of bread has, which is actually very very cheap. But when it comes to the transients, do you see if it becomes a little bit more interesting? So you have a useless like fashion store, where you have you want to store the keys for a certain amount of time? But do you want to be available for that? Because if you lose the key

and also translates to like session store, your keys are highly available. Define Restore for the opposition has very hard but you also want to ensure that the date is backed up at last week about to be cash, is interesting for the prospective. All customers the Florida. Cash the one of the key things that capacity casualties for, is to ensure that the back of store is not. So, what that means is that even if you wanted to know. Washing powder. I'm fixing to hit one of these cash and patterns. What are some of

the best practices that you should be following to ensure that you're running the memory Sorensen's? Optimally for your application working? So the first thing that I would like you to do is to understand and follow the memory management. Best practices for write, as one of the things I ask for is for memory management but has Amana service, we do a lot of things that are important for you to follow the best practices, especially ensure that your incense is wanting off

to me. Some of the metrics that you should be monitoring. Do we have the reporting metrics? In from our perspective. You should be monitoring if you're using them when it comes to men ratio, basically tells you how much system memory is getting used to know if your instant are your work. Phone is coughing. A lot of memories too brother. You're deployed. Is basically being used optimally for my data usage perspective. The next one is also very interesting

and something that's what happens behind the scenes as bad experience. So one of the things that you want to monitor is that feeds your incense getting into that condition or know, right? It could be that close where you actually allow you to I lost the election to ensure that if you have, you know, everything policy set where you expect that the keys are getting, if it's just rather tell you whether you're getting into a bath. So once you're done with your memory usage, best practices and the monitoring around. Then the next thing you want to do is based on the signals.

So like I mentioned earlier, this is your memory usage ratio is a very good indicator of how well in your system or incense is performing in terms of consumption you what you what you allow you to give them more overhead for the for the instant. This is super important especially when you are doing things like when you stay up Anson and Ashley is Double the memory that you're using defending on the right phrase on your on your instant, especially if you're in the right area, since we highly recommend

that you for operation. That's about right side thing. The instant. So just to recap Montefiore memory management. Best practices and white size based on the thresholds that you are the metrics are on diamond. And finally, I just wanted some general best practices that you should be following especially for the primary store you skis. So one of the things I mentioned earlier is that all you're going against a single. So what that means is that I'm heading to the limits

or right intensive, are you reading want to monitor your memory usage like and follow the best practices that I mentioned in a previous life? I know you said you can also determine whether you're hitting the limit if there's a lot of rice and you can basically running the scenario. And lastly, if you're using it as a prime restore, your absolutely should take a backup off your instant published a paper on how to schedule publishing our solution website using

their research. And they are thinking about your actions, depending on what the metrics. Are telling me that you're essentially being proactive, ensuring that the memories also more importantly, it is actually a suitable for you in some cases. Maurice oranges are not able to keep up, then you will have to share your memories store or application for multiple. The. Hopefully gives you a general understanding of what are some of the best practices that you should be following when using them?

The word that I just started Texarkana are some useful links that will help you understand a lot more about what I've just talked about to your backups all this together. Hopefully, it will give you a great start in, ER,

Cackle comments for the website

Buy this talk

Access to the talk “Optimally deploy an application cache with Memorystore”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Ticket

Get access to all videos “Google Cloud Next 2020”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Ticket

Similar talks

Sophie Hu
Data Management Specialist, Customer Engineering at Google
+ 1 speaker
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Gurmeet (GG) Goindi
Product Management at Google
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Sean-Michael Lewis
Senior Staff Software Engineer at Google
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Buy this video

Video

Access to the talk “Optimally deploy an application cache with Memorystore”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Conference Cast

With ConferenceCast.tv, you get access to our library of the world's best conference talks.

Conference Cast
635 conferences
26170 speakers
9693 hours of content