AWS Lambda needs to improve cold start time. Some AWS Lambda functions require a cold start, and if you need AWS Lambda to provide quick responses, you need some of the AWS Lambdas to be always on, which is risky. We need AWS Lambda's cold start time to be reduced so that we can use it much faster than now. We need a better way to handle the cold start. We should be able to start AWS Lambda much before in a predictable way instead of just calling and then having it start.
Other than memory limitations, the areas I would to see improved or enhanced in AWS Lambda are mainly concerning monitoring, as everything else is pretty good and fast. My concerns about monitoring in AWS Lambda are that the logs they provide are not easily understandable, and how they provide it makes it take time to find specific logs and that particular number itself.
Having a better preview would be helpful. Sometimes, if my Lambda code is too big, it can be inconvenient as I'm unable to see my code when it exceeds a certain size. AWS has a limit, like a three-megabyte limit, beyond which I cannot view or edit the code easily.
The runtime could be improved. There are certain use cases where I need a Lambda function to run longer. I don't know how many minutes it is, however, there's a threshold that I can't exceed. If that time was extendable, it would help in some areas, not all areas, yet some.
Desarrollador de Microsoft at Sinnvoll analítica empresarial
Real User
Top 10
Jan 8, 2025
The way Lambda works with HTTP headers is a bit odd. It needs some change as the configuration is somewhat hidden, so I don't really know what I'm changing. Managing the HTTP headers should be modified for clarity since, at least in my case, I found it challenging to configure and manage them initially. I would suggest that this is an area for improvement.
AWS Lambda is a bit difficult to set up if someone doesn't know how to code. It would be helpful to have a chatbot or resource that can guide users through the process, especially for those not proficient in coding. Setting up workflows that connect multiple AWS services can be challenging without guided support.
One area where AWS Lambda could be improved is the execution timeout limit, which is currently capped at 15 minutes. This is restrictive for tasks that require longer processing times, such as complex data analysis or large-scale batch processing. Additionally, the environment customization is limited, particularly with regard to file system access, persistent storage, and certain execution capabilities, which can be restrictive for more advanced use cases. Another area for improvement is the cold start latency, which can add significant overhead to the execution time, especially for applications with infrequent or unpredictable invocations. For future releases, it would be helpful if AWS Lambda allowed for longer execution times or offered an option for more flexible execution environments. Reducing cold start latency or offering features like provisioned concurrency with more granular control could also help improve the overall performance and responsiveness of Lambda functions.
AWS Lambda has a limitation where the execution time is capped at 15 minutes per task. Increasing this time would allow for handling heavier tasks more efficiently. While the documentation is satisfactory, beginners might encounter some challenges, so any improvements could be helpful.
One area of improvement is Lambda's runtime version management. It currently requires manual user maintenance to upgrade and evaluate, and an automated provision for this would be beneficial.
AWS Lambda functions have many versions. Sometimes, it takes a little time to load and run the commands in AWS Lambda. We can write anything as code, but the solution will not give proper error information. It is a little time-consuming to understand where the error came from.
They could add Copilot or their own suggestion tool, like CodeWhisperer, to the AWS IDE. The current IDE seems outdated and could be improved. Also, Lambda is restricted in terms of Internet access. You can't get public connectivity directly. You need to download packages manually elsewhere, zip the files, and add the layer manually. You can't install packages directly from the Internet, like PIP install Pandas.
We must at least know the basics of languages like Python or Java to implement automatic processes and daily schedules. We must learn the Lambda function and the languages to use the tool.
There's room for improvement in the testing setup. Attempting to test it through the AWS CLI also posed issues, and when trying to test the Lambda function on my local machine, I encountered difficulties in obtaining comprehensive logs or error messages to diagnose what might be going wrong.
Principal Solution Architect at a construction company with 51-200 employees
Real User
Jul 19, 2023
Memory limitation is one of the weaknesses of AWS Lambda and as a result, we have to use several Lambda, instead of just one. Recently, I met with an Amazon employee, who is responsible for Lambda as a product. It appears Amazon has some plans with Lambda, so I don’t have to add something to the additional features.
There were some timeout issues with AWS Lambda as the options provided didn't suit our business cases. In the next release, AWS Lambda should include integration with an IoT.
Senior ict specialist at Information& eGov Authority
Real User
Jan 6, 2023
My opinion is that the integration could be improved in this solution. We have had some difficulties integrating the EC2 module, but we found a solution for that by ourselves. I would say the integration needs to be improved.
I would like the layers to have a bigger volume. I would like to be able to add more. I don't want to be limited by the layer. In addition, writing temporary data to Lambda is not straightforward. Sometimes I use the S3 for that. I found it a little bit challenging.
Genesys Cloud Consultant / Software Engineer at Hightelecom
Real User
Nov 18, 2022
The 60 seconds limitation with the consumption of the service is really restrictive for a service and the solution can be improved by eliminating that.
What could be improved in AWS Lambda is a tricky question because I base the area for improvement on a specific matrix, for example, latency, so I'm still determining if I can be the judge on that. However, room for improvement could be when you're using AWS Lambda as a backend, it can be challenging to use it for monitoring. Monitoring is critical in development, and I don't have much expertise in the area, but you can use other services such as Xray. I found that monitoring on AWS Lambda is a challenge. The tool needs better monitoring. Another area for improvement in AWS Lambda is the cold start, where it takes some time to invoke a function the first time, but after that, invoking it becomes swift. Still, there's room for improvement in that AWS Lambda process. In the next release of AWS Lambda, I'd like AWS to improve monitoring so that I can monitor codes better.
There's room for improvement in the solution's warm start, which refers to the minimum time it takes to start up a Lambda function if you haven't been running it.
Lead Data Engineer at Seven Lakes Enterprises, Inc.
Real User
Aug 1, 2022
The execution time could be better. One of the major limitations is the time period because it doesn't give you more than seven minutes. So, before thinking about Lambda, you should think through your use case and ensure it's a good fit. Otherwise, you can use batch, step functions, or other methods. Reports and the monitoring board could also be improved in terms of alerts. The threshold alerts are there but can be improved. It takes some time to get used to these methods and get the hang of them.
I would like to see a find and replace function as part of Lambda's future releases. Currently, if we want to replace a code, we copy the code into Notepad, then find and replace it, and then copy that to Lambda. This would improve the editing function of the product. Lambda would benefit from a debugging feature as well. For example, if you want to debug code running in Python and deployed in Lambda, it would be beneficial to have a debugging feature.
Architect - Database Administration at Mitra Innovation
Real User
Top 5
Jun 3, 2022
From a very high-level perspective, it is a good solution, however, there are certain limitations. This becomes challenging for us. Specifically, a limitation is Lambda cannot continuously execute for more than 15 minutes. That could be one limitation. On an implementation level, there are certain limitations that I know about yet would not be the best person to explain as I am not technical. The timeout for execution can be increased so that if we have a lengthy process, it will not fail before their timeouts. We've had to revamp the way that it works due to that 15-minute timeout limitation.
Consultant at a educational organization with 11-50 employees
Real User
Jul 11, 2021
The product could make the process of integration easier. The could be more integration testing. The internal is great with Lambda, however, I have to make sure I have done my tests on a related machine before deploying.
The user-friendliness of the solution could be improved. If it was easy to run with the same function in other platforms, other environments, that would make it more portable. That would be really good. User-friendliness and portability will be the two factors that need the most improvement. The startup time sometimes needs to be faster, so that is one area of improvement. The startup time of each function can be slow. When it works the first time, it takes a little bit of time, so there's a minor delay. That could be improved. The support of additional languages would improve the solution.
Sr. Technology Specialist at a computer software company with 5,001-10,000 employees
Real User
Mar 10, 2021
Its price should be improved. Its pricing is on the higher side. I am not sure if it currently supports the Go language. If it doesn't support the Go language, they can introduce it.
IoT/AI/Enterprise Solutions Architect at Tech Data Corporation
Real User
Nov 29, 2020
I can only speak from the user experience. Salesforce integrates SharesPost efficiently. How? They say, "Okay, I invented another language called Apex. Forget about anything else. This is my language." The benefit of this language is that everything is simplified. Your system is super easy to maintain. But AWS then assures you that they are flexible, that they have a collection of 10 or 20 languages, and you just choose your environment and range. That's the reason I appreciate Salesforce. They always make things easier. They have their loop reasoning because they are a different kind of company. Microsoft and AWS really get the full spec. They want to own the business. But Salesforce data wants the simplest way. So, this is my understanding and unique experience. I think that perhaps Lambda could explore its functionality more.
The kind of scenarios we can leverage in Lambda can be difficult. For example, we have a lot of their servers that provide APIs. Before, the APIs were hosted by a server. We have identified that some of them can be hosted in Lambda. That's a challenge as we're yet to think of our scenario and match our requirements in order to really leverage Lambda. That's more on us than the solution itself. We need to better understand Lambda for different scenarios. We need some joint effort between Amazon and the users to have the users identify how they can really leverage Lambda. It's not about Lambda itself; it's about the practice, the guidance. There needs to be very good documentation. From the user perspective, what exists now is not always enough.
One area of improvement is to include support for more programming languages. AWS Lambda does not support a lot of programming languages. You have to write the Lambda functions in a certain programming language. We are using C++. My developer knows a couple of other languages. Python is his favorite language, but Python is not supported in AWS Lambda.
Assistant to Vice President at a consultancy with 10,001+ employees
Real User
Top 20
Apr 19, 2020
Lambda didn't use to support all the capabilities, if you wanted to initialize it before started serving the requests, it is not possible. In the most recent upgrades, they have added this feature. There is nothing missing. If there is something missing, AWS has other solutions that covers it. I look at Lambda as a piece of AWS, working in conjunction with other solutions.
AWS Platform Head & Chief Architect - CMA Unit at a tech services company with 10,001+ employees
Real User
Mar 3, 2020
There are certain limitations to some services in Lambda which will have to be overcome over time. Primarily, I think there are two aspects of the product that are in most need of improvement. One is, of course, the customization. The process of customizing should be simplified. The other is specifically in the area of integration. The configuration requirements for the services which would be desirable to integrate with Lambda are not always as easy to utilize as they should be. For example, the opportunities we have to integrate solutions with the world outside of the cloud are not well supported. These integrations are supported well inside the cloud but some clients need to support solutions outside the cloud. If somebody has that need we should be able to deploy integrations with products and services outside the cloud as well. So those are the two drawbacks and potential areas for improvement area which I would like to see corrected in AWS. Those features need to be updated because the current functionality of the product is not competitive enough today with other products in the category. I think with time they have to improve and add new features or offer other ways of integrating new services.
There are several things that can be improved, including: * It has limited access to disk space, as it does not have full access to underline VM * Lamba functions have cold-starts that can cause some delay * It is hard to troubleshoot and monitor Lambda functions
AWS Lambda is a compute service that lets you run code without provisioning or managing servers. AWS Lambda executes your code only when needed and scales automatically, from a few requests per day to thousands per second. You pay only for the compute time you consume - there is no charge when your code is not running. With AWS Lambda, you can run code for virtually any type of application or backend service - all with zero administration. AWS Lambda runs your code on a high-availability...
AWS Lambda needs to improve cold start time. Some AWS Lambda functions require a cold start, and if you need AWS Lambda to provide quick responses, you need some of the AWS Lambdas to be always on, which is risky. We need AWS Lambda's cold start time to be reduced so that we can use it much faster than now. We need a better way to handle the cold start. We should be able to start AWS Lambda much before in a predictable way instead of just calling and then having it start.
Other than memory limitations, the areas I would to see improved or enhanced in AWS Lambda are mainly concerning monitoring, as everything else is pretty good and fast. My concerns about monitoring in AWS Lambda are that the logs they provide are not easily understandable, and how they provide it makes it take time to find specific logs and that particular number itself.
Currently, I don't have any specific suggestions for improvements. However, enhancements could be introduced to better satisfy customer requirements.
Having a better preview would be helpful. Sometimes, if my Lambda code is too big, it can be inconvenient as I'm unable to see my code when it exceeds a certain size. AWS has a limit, like a three-megabyte limit, beyond which I cannot view or edit the code easily.
The runtime could be improved. There are certain use cases where I need a Lambda function to run longer. I don't know how many minutes it is, however, there's a threshold that I can't exceed. If that time was extendable, it would help in some areas, not all areas, yet some.
The way Lambda works with HTTP headers is a bit odd. It needs some change as the configuration is somewhat hidden, so I don't really know what I'm changing. Managing the HTTP headers should be modified for clarity since, at least in my case, I found it challenging to configure and manage them initially. I would suggest that this is an area for improvement.
AWS Lambda is a bit difficult to set up if someone doesn't know how to code. It would be helpful to have a chatbot or resource that can guide users through the process, especially for those not proficient in coding. Setting up workflows that connect multiple AWS services can be challenging without guided support.
One area where AWS Lambda could be improved is the execution timeout limit, which is currently capped at 15 minutes. This is restrictive for tasks that require longer processing times, such as complex data analysis or large-scale batch processing. Additionally, the environment customization is limited, particularly with regard to file system access, persistent storage, and certain execution capabilities, which can be restrictive for more advanced use cases. Another area for improvement is the cold start latency, which can add significant overhead to the execution time, especially for applications with infrequent or unpredictable invocations. For future releases, it would be helpful if AWS Lambda allowed for longer execution times or offered an option for more flexible execution environments. Reducing cold start latency or offering features like provisioned concurrency with more granular control could also help improve the overall performance and responsiveness of Lambda functions.
AWS Lambda has a limitation where the execution time is capped at 15 minutes per task. Increasing this time would allow for handling heavier tasks more efficiently. While the documentation is satisfactory, beginners might encounter some challenges, so any improvements could be helpful.
A very minor improvement would be to simplify the instructions on setting a trigger, as I had to read through them multiple times at the start.
One area of improvement is Lambda's runtime version management. It currently requires manual user maintenance to upgrade and evaluate, and an automated provision for this would be beneficial.
AWS Lambda functions have many versions. Sometimes, it takes a little time to load and run the commands in AWS Lambda. We can write anything as code, but the solution will not give proper error information. It is a little time-consuming to understand where the error came from.
They could add Copilot or their own suggestion tool, like CodeWhisperer, to the AWS IDE. The current IDE seems outdated and could be improved. Also, Lambda is restricted in terms of Internet access. You can't get public connectivity directly. You need to download packages manually elsewhere, zip the files, and add the layer manually. You can't install packages directly from the Internet, like PIP install Pandas.
We must at least know the basics of languages like Python or Java to implement automatic processes and daily schedules. We must learn the Lambda function and the languages to use the tool.
AWS Lambda should support additional languages.
We face some problems with the event-driven execution model.
There's room for improvement in the testing setup. Attempting to test it through the AWS CLI also posed issues, and when trying to test the Lambda function on my local machine, I encountered difficulties in obtaining comprehensive logs or error messages to diagnose what might be going wrong.
Memory limitation is one of the weaknesses of AWS Lambda and as a result, we have to use several Lambda, instead of just one. Recently, I met with an Amazon employee, who is responsible for Lambda as a product. It appears Amazon has some plans with Lambda, so I don’t have to add something to the additional features.
AWS Lambda needs to improve its stability.
I want to see support for longer applications. I need the 15-minute time-out window to improve.
The tool changes its UI every month which is very frustrating for me. I don’t know why AWS keeps changing the UI. They can’t stick to a specific one.
There were some timeout issues with AWS Lambda as the options provided didn't suit our business cases. In the next release, AWS Lambda should include integration with an IoT.
My opinion is that the integration could be improved in this solution. We have had some difficulties integrating the EC2 module, but we found a solution for that by ourselves. I would say the integration needs to be improved.
I would like the layers to have a bigger volume. I would like to be able to add more. I don't want to be limited by the layer. In addition, writing temporary data to Lambda is not straightforward. Sometimes I use the S3 for that. I found it a little bit challenging.
The 60 seconds limitation with the consumption of the service is really restrictive for a service and the solution can be improved by eliminating that.
It can be a bit difficult to switch between accounts when creating services for customers.
There are sometimes issues following an update.
What could be improved in AWS Lambda is a tricky question because I base the area for improvement on a specific matrix, for example, latency, so I'm still determining if I can be the judge on that. However, room for improvement could be when you're using AWS Lambda as a backend, it can be challenging to use it for monitoring. Monitoring is critical in development, and I don't have much expertise in the area, but you can use other services such as Xray. I found that monitoring on AWS Lambda is a challenge. The tool needs better monitoring. Another area for improvement in AWS Lambda is the cold start, where it takes some time to invoke a function the first time, but after that, invoking it becomes swift. Still, there's room for improvement in that AWS Lambda process. In the next release of AWS Lambda, I'd like AWS to improve monitoring so that I can monitor codes better.
There's room for improvement in the solution's warm start, which refers to the minimum time it takes to start up a Lambda function if you haven't been running it.
The overall performance of this solution could be improved. We would also like to have better integration with other AWS features.
AWS Lambda can improve its file system-based sharing capabilities and restrictions.
The execution time could be better. One of the major limitations is the time period because it doesn't give you more than seven minutes. So, before thinking about Lambda, you should think through your use case and ensure it's a good fit. Otherwise, you can use batch, step functions, or other methods. Reports and the monitoring board could also be improved in terms of alerts. The threshold alerts are there but can be improved. It takes some time to get used to these methods and get the hang of them.
AWS Lambda has some size limitations in the code line, you can't do a couple of functions to do the job.
I would like to see a find and replace function as part of Lambda's future releases. Currently, if we want to replace a code, we copy the code into Notepad, then find and replace it, and then copy that to Lambda. This would improve the editing function of the product. Lambda would benefit from a debugging feature as well. For example, if you want to debug code running in Python and deployed in Lambda, it would be beneficial to have a debugging feature.
AWS Lambda's GUI could be improved with a twist or tweak in its look and feel to make it more impressive.
From a very high-level perspective, it is a good solution, however, there are certain limitations. This becomes challenging for us. Specifically, a limitation is Lambda cannot continuously execute for more than 15 minutes. That could be one limitation. On an implementation level, there are certain limitations that I know about yet would not be the best person to explain as I am not technical. The timeout for execution can be increased so that if we have a lengthy process, it will not fail before their timeouts. We've had to revamp the way that it works due to that 15-minute timeout limitation.
We don't have the inbuilt modules in AWS Lambda. If more modules were built into or integrated with AWS Lambda, that would help developers to code.
AWS Lambda could improve by having no-code or low-code options because currently, you need to be able to write code well to use it.
AWS Lambda could be improved with better stability.
I wish to see better execution time in the next release.
The automation with other Amazon products could be better.
I believe Lambda security needs to be improved.
The product could make the process of integration easier. The could be more integration testing. The internal is great with Lambda, however, I have to make sure I have done my tests on a related machine before deploying.
There is room for improvement in user-friendliness. When comparing this solution to others it is not as user-friendly.
The user-friendliness of the solution could be improved. If it was easy to run with the same function in other platforms, other environments, that would make it more portable. That would be really good. User-friendliness and portability will be the two factors that need the most improvement. The startup time sometimes needs to be faster, so that is one area of improvement. The startup time of each function can be slow. When it works the first time, it takes a little bit of time, so there's a minor delay. That could be improved. The support of additional languages would improve the solution.
Its price should be improved. Its pricing is on the higher side. I am not sure if it currently supports the Go language. If it doesn't support the Go language, they can introduce it.
I can only speak from the user experience. Salesforce integrates SharesPost efficiently. How? They say, "Okay, I invented another language called Apex. Forget about anything else. This is my language." The benefit of this language is that everything is simplified. Your system is super easy to maintain. But AWS then assures you that they are flexible, that they have a collection of 10 or 20 languages, and you just choose your environment and range. That's the reason I appreciate Salesforce. They always make things easier. They have their loop reasoning because they are a different kind of company. Microsoft and AWS really get the full spec. They want to own the business. But Salesforce data wants the simplest way. So, this is my understanding and unique experience. I think that perhaps Lambda could explore its functionality more.
The kind of scenarios we can leverage in Lambda can be difficult. For example, we have a lot of their servers that provide APIs. Before, the APIs were hosted by a server. We have identified that some of them can be hosted in Lambda. That's a challenge as we're yet to think of our scenario and match our requirements in order to really leverage Lambda. That's more on us than the solution itself. We need to better understand Lambda for different scenarios. We need some joint effort between Amazon and the users to have the users identify how they can really leverage Lambda. It's not about Lambda itself; it's about the practice, the guidance. There needs to be very good documentation. From the user perspective, what exists now is not always enough.
One area of improvement is to include support for more programming languages. AWS Lambda does not support a lot of programming languages. You have to write the Lambda functions in a certain programming language. We are using C++. My developer knows a couple of other languages. Python is his favorite language, but Python is not supported in AWS Lambda.
Lambda didn't use to support all the capabilities, if you wanted to initialize it before started serving the requests, it is not possible. In the most recent upgrades, they have added this feature. There is nothing missing. If there is something missing, AWS has other solutions that covers it. I look at Lambda as a piece of AWS, working in conjunction with other solutions.
There are certain limitations to some services in Lambda which will have to be overcome over time. Primarily, I think there are two aspects of the product that are in most need of improvement. One is, of course, the customization. The process of customizing should be simplified. The other is specifically in the area of integration. The configuration requirements for the services which would be desirable to integrate with Lambda are not always as easy to utilize as they should be. For example, the opportunities we have to integrate solutions with the world outside of the cloud are not well supported. These integrations are supported well inside the cloud but some clients need to support solutions outside the cloud. If somebody has that need we should be able to deploy integrations with products and services outside the cloud as well. So those are the two drawbacks and potential areas for improvement area which I would like to see corrected in AWS. Those features need to be updated because the current functionality of the product is not competitive enough today with other products in the category. I think with time they have to improve and add new features or offer other ways of integrating new services.
The running time of AWS Lambda runs fine. It takes around five minutes but it would be great if that time could be extended.
The security needs to be improved.
There are several things that can be improved, including: * It has limited access to disk space, as it does not have full access to underline VM * Lamba functions have cold-starts that can cause some delay * It is hard to troubleshoot and monitor Lambda functions
I would like to see some better integration with other providers, like Cohesity, Druva, and others. I also think the Lambda interface could be better.