Accelerate your cloud business with AWS built-in partner solutions

Anthony Palladino: I'm Anthony Palladino, uh global lead for aws built in competency.

Kisho Rum: Hi, this is kisho rum. I'm a solution architect focused on building the solutions.

Anthony: So what is eight of those built in? I'll just give a brief overview before we kind of get into the slides, but built in is really designed to be a way to lift the operational burden off a customer's shoulders uh when they're deploying first party software and partner third party software. So just a way to simplify for our customers.

So if i ask for a show of hands, which i won't, but i was gonna say, you know, how many of you have extra time resources and skill, probably not many and often that's really the reason why you're in cloud in the first place, right? Because cloud in and in and of itself drives efficiency for our customers.

Um so what built in does is it's trying to give some of that back to our customers and leveraging our third party partners to do it. So if you look at this quote from mckinsey, um basically, they're saying it's really important to get the building blocks of cloud, right? When you migrate to cloud or if you're already in existing cloud and you're hitting that cloud 2.0 and you need to modernize it's really important to get it right, because ultimately that's going to drive growth and scale for your business.

Uh the other thing it does is you want to make sure you get those building blocks of security, right? Um and eight of a built in focuses in on that as well.

So a couple of things uh aws built in started as a pilot this year, we engaged just a handful of our isv partners uh for those of you who don't know isv kind of a aws term stands for independent software vendor. Uh and they, the folks that have s a solutions on marketplace, uh you could go there, you could find them and you know, get that innovation that you need.

So aws provides a lot of that innovation, but we need help, just like other people need help and we often look to our partners to help us through that. Uh we have one of the early adopters of aws milton sitting right here in the audience. So, uh michael musselman from lacework, lacework was one of our pilot partners.

Uh and they helped us innovate and uh and then drive this uh drive this initiative. So what does aws built in do? So it's all about getting. So one of the ways i i describe it is a lot of times, uh, one of our isv partners is in the program and in a way it automates their deployment guide because a lot of times we don't, we don't give them really prescriptive guidance. That's not what they would do on their own in the first place. It just gets you the customer there faster.

So i'll take lacework as an example. Lacework when you have a lacework built in and you enable that in your environment, not only does it leverage control tower and organizations identity center, but it'll uh it'll integrate guard duty, security hub and org level trail.

Uh and all that could be deployed in the space of a single day. So again, really drives business efficiency uh for you as a customer.

Um so we do have a couple of quotes here and uh from some of the early adopters, this one from uh trend, what susanna is really trying to say when she says it's going to change the way they engage customers and partners when she mentions channel partners a lot of times she's talking about systems integrators.

So systems integrators often help our isvs or help our customers deploy our isv software. So think about if you have an aws built in and you're engaged with trend micro, when you go to an si partner, uh those building blocks are kind of going to be set now, we can really focus on like the fine tuning of the trend micro product and the same with lace work.

So what michael is saying is, hey, is it really, it's an easy button for aws. So aws can be complicated. Uh cloud in general can be complicated and you want to get the building blocks right? Uh aws built in is going to help you do that. And with that, we can kind of dive deep and really get into like the technical aspect of it and uh keep sure.

Kisho: All right, hey, everyone, hope you're having a great time at reinvents so far. This is kisha win. I'm a solution architect focused on the build part of aws building. So i work with partners designing and building these solutions.

So if you are a techie like me wondering about how these solutions are built and what mechanisms are in place to ensure the quality of this of solutions are maintained. I'm going to talk about it in the next five minutes.

I understand five minutes is too little for a conversation like this. So we are at booth number 314. Feel free to stop by if you have any questions or would like to talk about your particular use cases. Me and anthony will be available there and happy to talk about uh in detail.

So as anthony mentioned, like we spoke to several aws customers and what we learned is our customers rely on aws and a rich set of partner ecosystem to power their business and scale their business needs as they grow.

So for that, like, so these aws partners, they rely on the foundational services, uh the insights that are provided by the foundational services like trails from aws cloud trail or metrics from cloudwatch or findings from the guard duty and so on. So on, right.

So having for the customers who choose these third party tools, they want to assure like the tools that they are picking are well integrated with aws services and also have they are cost optimized and they are they scale with their business needs this.

So this is the reason why we partnered with aws partners and build a solutions where we provide a modular code repos for the aws native services.

So for partners use the aws native services, the model code we provide to enable these services and partners configure their additional integrations on top of it, right?

So they bundle this together like for partners, they don't need to stay on top of what's a ws is iterating on as, as the service grows, that is a control that's managed by aws and partners can leverage what is being built by aws and it has a identical deployment of native services across the platform.

So once partner build these solutions together, like with, with native services and the partner configuration, uh they have they submit the code and we go through an auto automatic validation process, right?

So i will talk in detail about this validation process. But what we look for here is an a, a group of uh comm and a lightweight manual process to make sure the code that not only while building phase like the ongoing check ins that comes into the code is verified deployed and validated.

And then we publish it to the targets that customers would consume from. So when we qualify a or identify a solution integration with partners, so we build, we create a github repo with a sample repos is like a common baseline templates for the partners.

So this includes some of these template that's being used. And then we also dedicate a my account environment for the customers, right?

So this this particular solution will will be leveraging this environment that we assign to validate as an ongoing process. So once the partner builds these solutions and they submit the code to the bay, there is an automation that kicks in and performs certain tests, right?

So these tests include validating for looking for any known security issues or vulnerabilities or and or are there any deviations to the best practices of the services that's being used?

So it verified those information and once that is true that it will provide the information to it, partner solution architect who will review the code and say trigger the functional test.

So these functional tests are basically going to deploy these confirmation templates under various scenarios into the environment that is provided, right?

So this would cover, make sure like yeah, the the uh any changes that comes to these solutions are vetted much before and it will pass on the information back to the uh partner solution architect with an approval and psa will review this code and then approve, provide the second approval once he is satisfied with the info with with the check in and that will trigger the publication phase, right?

So the publishing is where it is going to make the templates available to the end point where from where the customers would consume it. So this could be an aws marketplace or this could be a github repo itself or it could be from within the service catalog uh and made it made that available to the partner to the customer.

So you may have seen like here we are, we are doing a bunch of automation here and we have some manual tests as well. So you might be wondering why we are doing this, right?

So, so while we rely on automation testing for the scalability purposes, we do want to ensure not only the quality of the code, but also we wanted to learn from uh the the check ins that comes in as we build and we are having a manual checks to make sure like the code is the code we are getting is behaving as expected and we will its rate on the automation, right.

So our end goal is to have everything automated but we will have a lightweight manual test as we, as we start building these environments more.

All right. So with that, you might be wondering like uh who are the partners that we worked on? Of course, we talked about list work, but we worked with a set of partners uh for the launch and i will uh have anthony talk about those details.

Anthony: Thank you, kur. So these are built in partners. They all participated in the 2023 pilot. We started working with them uh back in q one. And they were the first ones to complete the co build.

Um super happy, super thankful to work with all of them. And just, you know, just as a call out, uh you'll see some of them will have multiple products, right? Not everyone just has like a, like a, like a single state product.

Um so the built in gets applied to a specific product and you know, so it's, it's very targeted. Um and then when you, when a customer uh enables that product, uh that a b i code uh will enable that product. And then if you're a customer and you wanna learn more about these uh uh partners, you could find them uh on aws marketplace.

Even if you just search uh aws uh built in on marketplace, you'll get to a page that has these partner solutions. Uh and if you're um if, if you're a partner that wants to learn more about actually joining the program. In that case, uh you can just contact and reach out to your pdm and your pdm will uh help you kind of, you know, navigate that.

So with that, uh here's a way to kind of find uh some of the information we talked about and again, reminder, we're at booth 314. Uh really appreciate, appreciate your time. Thank you.

Thanks to everyone and feel free to connect to us on uh on linkedin if you like. Thank you.

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
### 回答1: ClassSR是一种通过数据特征来加速超分辨率网络的通用框架。该框架可以根据不同的数据特征,自动选择合适的网络结构和超分辨率算法,从而提高超分辨率的效率和精度。同时,ClassSR还可以通过数据增强和模型蒸馏等技术,进一步提升超分辨率的性能。总之,ClassSR是一种非常实用的超分辨率技术,可以广泛应用于图像和视频处理领域。 ### 回答2: classsr: a general framework to accelerate super-resolution networks by data characteristic 是一个用于加速超分辨率网络的通用框架,通过对数据特征进行分析和处理,提高了超分辨率网络的训练效率和超分辨率图像的质量。 在超分辨率图像的生成中,一般采用深度学习的方法。但是传统的超分辨率网络存在训练速度慢、参数量多、生成的图像模糊等问题。classsr 认为这些问题是因为原始数据的特征并没有在训练过程中得到充分利用所导致的。 为了解决这些问题,classsr 提出了一种新的训练框架,它能够利用数据特征来加速训练过程和提高超分辨率图像的质量。具体来说,classsr 框架主要包含以下两个部分: 1. 特征提取的方法:classsr 采用了自适应卷积核技术,可以根据原始数据的纹理和结构特征,动态生成不同大小和形状的卷积核,从而提取更加准确的特征信息。 2. 数据特征的建模:classsr 通过分析数据的结构和特点,建立了一种数据特征模型,可以自动学习数据的统计特性。然后,利用这些数据特征来指导网络的训练,使得网络更快更准确地收敛。 总体来说,classsr 框架可以有效提高超分辨率网络的训练效率和超分辨率图像的质量。未来,该框架还能够为其他图像处理任务提供新的思路和方法。 ### 回答3: classsr是一种通用的框架,用于通过数据特性加速超分辨率网络。这种框架旨在提高超分辨率网络的训练速度和效果,并为图像和视频处理领域的任务提供更好的解决方案。 classsr的设计基于三个核心概念:数据特性,特征捕捉和解码器设计。首先,该框架通过对数据进行分析,确定了数据的特性。然后,特征捕捉模块通过特征检测和分类来提取输入图像的特征。最后,解码器根据提供的特征对图像进行重建。 classsr可以加速超分辨率网络的训练速度。该框架使用了轻量级的网络结构和高效的损失函数,使得训练速度比传统的超分辨率网络更快。同时,classsr还可以提高超分辨率网络的效果。该框架可以通过对数据特征的分析来优化网络结构,提高网络的性能和稳定性。 除此以外,classsr还可以为图像和视频处理领域的任务提供更好的解决方案。 classsr可以处理各种不同类型的图像和视频,并为各种应用场景(例如图像增强、视频压缩等)提供专门的解决方案。 综上所述,classsr是一种通用的框架,可以加速超分辨率网络的训练速度并提高网络的效果。该框架还可以为图像和视频处理领域的任务提供更好的解决方案。

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值