This article authorizes the official Amazon Cloud Technology article to forward and rewrite the rights, including but not limited to Amazon Cloud Technology official channels such as Amazon Cloud Technology Developer Community, Zhihu, self-media platforms, third-party developer media, etc.
expected
On September 25, 2023, Amazon announced a formal strategic cooperation with Anthropic, combining both parties’ advanced technologies and expertise in the field of safer generative AI to accelerate the development of Anthropic’s future basic models and make them widely available. Available to Amazon Cloud Technology customers.
As 2023 comes to an end, Amazon Cloud Technology announced a strategic cooperation with Anthropic at the end of September. This cooperation is Amazon Cloud Technology’s continued expansion of its generative AI “three-layer architecture”. Latest launch of unique services. At the bottom level, Amazon Cloud Technology continues to provide computing instances from NVIDIA, as well as Amazon Cloud Technology's self-developed chips, including Amazon Trainium for AI training and Amazon Inferentia instances for AI inference.
At the mid-tier, Amazon Cloud Technologies focuses on providing customers with the most popular base models from multiple leading model vendors. In addition, customers can customize these models to ensure the privacy and security of their data while seamlessly integrating with other Amazon Cloud workloads, all powered by Amazon Bedrock, a new Amazon Cloud service. Starting today, customers will have early access to the ability to customize Anthropic models, create their own private models with their own proprietary data, and use fine-tuning capabilities through Amazon Bedrock’s self-service.
At the top level, Amazon Cloud Technology provides customers with generative AI applications and services, such as the powerful AI programming assistant Amazon CodeWhisperer, which can generate code writing suggestions directly in the code editor. This improves the efficiency of developers writing code.
At this point, what we can already see is Amazon Cloud Technology’s long-term layout in AI, which has also laid a good foundation for subsequent development.
Amazon Q
After the launch of Amazon CodeWhisperer, expectations were raised for some of the functions of Amazon CodeWhisperer. As expected, it was launched in the new AI at the 2023 Amazon Cloud Technology re:Invent conference. Generative Assistant—Amazon Q
Amazon Q helps you quickly get relevant answers to pressing questions, solve problems, and generate content using data and expertise from your company's information repositories, code, and enterprise systems and take action. When you chat with Amazon Q, it provides instantly relevant information and advice to help simplify tasks, speed up decision-making, and help inspire creativity and innovation at work.
If there is a question, there must be an answer, **Amazon Q:** A new assistant based on generative artificial intelligence (AI), designed to assist work, can target Tailored to your customers' business, support for developers and IT professionals is available across multiple areas of AWS, giving you quick access to answers and ideas no matter where you work
Offered Multiple features and usage scenarios
With Amazon Q, AI experts are available to answer questions, code faster, troubleshoot issues, optimize workloads, and even help you write new features. These capabilities simplify all phases of building applications on AWS.
If you need additional assistance, Amazon Q also allows you to interact with an AWS Support agent directly from the Q interface, eliminating any pain points in the customer self-service experience. Integration with AWS Support is available in the console and provides benefits included in AWS Support plans.
Video demonstration
Amazon Cloud Technology re:Invent 2023 product experience: Amazon Cloud Technology product application practice, the best product Amazon Q, your AI assistant
Practice recurrence
In the official blog post, many usage methods are given, such as conversational Q&A, troubleshooting, code conversion and upgrade applications, etc. These are extremely practical functions, so let’s use these answers to reproduce the official cases.
Conversational Q&A
I believe everyone has used many generative AI assistants, such as OpenAI's ChatGPT, Google's Bard, Alibaba's Tongyi Qianwen, Baidu's Wenxinyiyan, etc. These are all conversational question and answer methods, so do you have any? Have you ever used an AI assistant on a cloud platform? You can talk to AI immediately directly on the console, so Amazon Q (preview version) of Amazon Cloud Technology is an AI assistant you absolutely cannot miss. It has professional knowledge of Amazon Cloud Technology and can help you when you are confused. , get fast and correct help.
Entry-level Q&A:
Simple questions and answers on basic knowledge:
Q：Hi,I want create a linux system instance to host my website,I should choose ec2 or lightsail?
A：Here are the key factors to consider when choosing between Amazon EC2 and Lightsail for hosting your website:
- Lightsail is easier to use and manage if you have a simpler workload like a basic website or app. It provides pre-configured options that can be set up with just a few clicks.
- EC2 gives you more flexibility and control over your server configuration but requires more devops knowledge. You have to manage components like networking and security groups yourself.
- Lightsail instances have fixed configurations while EC2 allows you to choose the exact compute, memory, storage and networking capabilities you need.
- The Lightsail interface provides an all-in-one dashboard for managing your resources while EC2 requires using different AWS console pages for different tasks.
- Basic Lightsail plans have all the necessary components included in a monthly flat fee. With EC2 you pay separately for individual resources like instances, storage, networking etc.
For a simple website that doesn't have demanding technical requirements, Lightsail would be easier to get started with. But for more complex or resource-intensive workloads, EC2 gives you more power and customization options at the cost of requiring more devops expertise.
Sources
[1] [What are the differences between Amazon EC2 and Amazon Lightsail?] (https://repost.aws/knowledge-center/lightsail-differences-from-ec2)
Professional knowledge Q&A
Q:If routes are advertised outside of AWS network, will this ASN be
pre-pend to the AS-path attribute of a route? If routes are advertised inside
AWS network, will this ASN be pre-pend to the AS-path attribute of a route?
A:
If routes are advertised outside of the AWS network, the ASN associated with the AWS Direct Connect connection or AWS Transit Gateway will not be prepended to the AS-path attribute of the route. This is because AWS does not actually handle routing or see the traffic in this case.
If routes are advertised within the AWS network (e.g. between a VPC and a Direct Connect gateway or Transit Gateway), the ASN will not be prepended to the AS-path either. AWS uses other mechanisms like local preference to determine the best path within AWS without modifying the AS-path.
The AS-path is generally not modified by AWS, but customers can influence route selection and traffic flow within AWS by configuring:
1. Local preference BGP communities
2. Advertising routes with different AS-path lengths
3. Configuring the AS-path attribute for interfaces in the same region
For more details on routing and BGP configurations within AWS, please refer to the AWS documentation pages on Direct Connect, Transit Gateway and VPC routing. The documentation provides the most accurate and up-to-date information.
Sources
[1] [Using AS-Path attribute to influence routing over multiple regions] (https://repost.aws/questions/QUPLkeHS0oT5qkHJv0d_9owA/using-as-path-attribute-to-influence-routing-over-multiple-regions)
[2] [AWS Transit Gateway ASN visibility ] (https://repost.aws/questions/QUbFjl8tPNS7uwBFMtQGdJFw/aws-transit-gateway-asn-visibility)
[3] [How Amazon Route 53 uses EDNS0 to estimate the location of a user - Amazon Route 53] (https://docs.aws.amazon.com/Route53/latest/DeveloperGuide/routing-policy-edns0.html)
EC2 instance selection optimization
The author has also experienced this function (Be careful to use the English interface). From a novice's perspective, this is indeed quite good. Yes, after all, Amazon Cloud Technology has too many EC2 instance families. Only by using the optimal instance will you gain advantages in terms of cost performance.
Then let’s use the default option WebHost to take a look at a suggestion from Amazon Q
For example, if we want to consider a price/performance ratio, then it is “Price/Performance”
What you can see is that almost all the recommended models are new models, such as M7, C7, and T4g, which happens to be a principle of Amazon Cloud Technology recommending users to use newer services.
troubleshooting
In fact, there can be several cases in this area,
- For example, if we give a user/role and restricted permissions, let it perform restricted operations.
- Official example, EC2 server failure
Because time is tight, I will only deal with the problem of EC2 network failure, and also add a little more difficulty.
EC2 network troubleshooting
Preset an EC2 instance. There are problems with the security group and routing table corresponding to the instance. See if Amazon Q can analyze it.
Prompt for security group repair: Then we will modify the security group to normal state and ask Amazon Q again
Asking again, it seems that we are just advised to check the security group. Here, obstacles are deliberately set up to "destroy routing". It seems that it is not recognized.
IDE integration dialog
I don’t know when Amazon Q has been in my VSCode. Looking at the official example, I installed AWS Toolkit, so Amazon Q naturally appeared in the function bar on the left side of my VSCode.
To use Amazon Q, you must log in to Amazon Builder before you can use Amazon Q in the IDE. At the same time, you can also describe the code we write in VScode and tell what our code is used for. Regarding the recent rise of "defensive code" And "ancestral shit mountain code", I think it is a good experience, and it is not so complicated to understand and analyze the code.
Summarize
After overall experience, Amazon Q's performance is quite satisfactory. It is indeed very familiar with the professional knowledge of Amazon cloud technology.
- It is at your fingertips and can be easily integrated into platforms and IDEs.
- Strong professionalism and accurate answers to professional knowledge about Amazon cloud technology
- A good code assistant, combined with Amazon CodeWhisperer, should have miraculous effects when writing code
Looking forward to improvements
- Network troubleshooting should be deeper than the appearance of the security group. The network itself is a very complex thing.
- Looking forward to launching a multi-language version, after all, the language proficiency of developers varies.