External reviews
External reviews are not included in the AWS star rating for the product.
CAST AI on Azure for larage scale spark workload
What do you like best about the product?
value, ease of integration, the ability to have fine-grained control over what and how we optimize
What do you dislike about the product?
That i did not use it before... (-:
but to be honest, It's one out of 3 or 4 tools that, in 20+ years of career, I'm struggling to find something to dislike
but to be honest, It's one out of 3 or 4 tools that, in 20+ years of career, I'm struggling to find something to dislike
What problems is the product solving and how is that benefiting you?
cloud cost... that is the main driver but we also have more benefit such as cost insight and perf
- Leave a Comment |
- Mark review as helpful
EC2 Worker Management Automated
What do you like best about the product?
Fairly simple to install, we used it's Terraform provider -- and the team onboarding with us actively addressed some minor issues from the provider on the fly (literal next day solutions +1). Initially i was pretty skeptical and nervous about handing over scaling of EC2 to a third party - but after it was in play we saw DRAMATIC savings wherever it was installed with our many EKS clusters, and little to no impact to services hosted inside of them. The impacts we had seen were from our own installations for PVC management, that we have since rectified and its been resilient since. Totally worth checking out, highly recommended.
What do you dislike about the product?
Workload optimization seems to strip out the CPU limits, and i dont see a means in the settings for it to brick that behavior. There are annoations etc - but would prefer something also to be globally accessible to prevent that. Additionally, node sizes work fairly well - but with worker consolidation we had some workers end up with 90+ pods on them -- and with our stack the resources aren't totally defined yet causing impact to other pods on the same ec2. Restricting total IP count as an optional setting would be nice.
Problems we experienced were almost entirely due to negligence from our internal staff with kubernetes manifests however -- If those were configured properly it wouldn't even be mentioned.
Problems we experienced were almost entirely due to negligence from our internal staff with kubernetes manifests however -- If those were configured properly it wouldn't even be mentioned.
What problems is the product solving and how is that benefiting you?
Instead of static set workers or the generic cluster-autoscaler for kubernetes hosted in EKS, it provides a way to consolidate pods to ec2 so it can be more compacted. This ultimately saved us a ton of money overall.
Huge savings on our AWS cloud expenses!
What do you like best about the product?
Cast does a great job at optimizing the workload, thus reducing the cloud costs. we were pretty optimized already and it still saved us around 30-40% of the total costs(!)
The team is very engaged and really care about our success. they are always there to answer questions and do deep dives when changes are made
Also they give very good transparency so you know where you stand
And I loved the onboarding process where they monitor your clusters and tell you how much money you could have spent, which was pretty accurate when we connected them. it was a smooth process, so implementation was really painless
The team is very engaged and really care about our success. they are always there to answer questions and do deep dives when changes are made
Also they give very good transparency so you know where you stand
And I loved the onboarding process where they monitor your clusters and tell you how much money you could have spent, which was pretty accurate when we connected them. it was a smooth process, so implementation was really painless
What do you dislike about the product?
Nothing specfic I dislike, there are always things to improve, but generally - I like the product and I like them as partners
What problems is the product solving and how is that benefiting you?
We have millions of K8S jobs running weekly. We were using the AWS scaler and we had a feeling we were paying too much for what we were running. One of our DevOps engineers thought we really need to try a solution and recommended we tried cast. from the moment we started using them we say 30-50% savings on clusters instantly.
The bin packing and better orchestration of jobs together to a great job at optimizing the use of compute resources and really saves you money
The only thing we regretted was that we waited so long
The bin packing and better orchestration of jobs together to a great job at optimizing the use of compute resources and really saves you money
The only thing we regretted was that we waited so long
A Kubernetes automation platform we were missing
What do you like best about the product?
So we had issues in a small team managing a couple of clusters in different environment and managing them in cost effective way. We appreciate the ease of use and the efficiency gains CAST provides, allowing teams to focus on development rather than infrastructure management. We also significantly reduced cloud costs for AKS in Azure. CAST is helping for managing cluster autoscaling and spawn optimal nodes. Also additional features like security scanning and giving potential vulnerabilities in our images helps a lot. The simplicity of installation of the platform, cost optimization ,security features , bin packing, cost tracking, workloads right sizing and many more makes the platform really brilliant. Also helping us to easily upgrade k8s from version to version. Even the features that we potenatially might need and are not yet implemented they listen and plan some feature request for them.
Last but not least is the awesome fast and accurate support from CAST team in case of issues.
Last but not least is the awesome fast and accurate support from CAST team in case of issues.
What do you dislike about the product?
I dont have any significant complains about CAST AI at this moment.
What problems is the product solving and how is that benefiting you?
Recommendations and cost optimization. Reducing the cost of AKS clusters.
Automation bean packing and ease of upgrade for k8s versions.
Workloads right sizing.
Reight sizing of the nodes and especially spot instances.
Responsiveness of CAST AI's support team in case of issues.
Detailed usage patterns, allowing teams to make informed decisions about optimizations.
Security image scanning feature for Vulnerabilities of the images.
Automation bean packing and ease of upgrade for k8s versions.
Workloads right sizing.
Reight sizing of the nodes and especially spot instances.
Responsiveness of CAST AI's support team in case of issues.
Detailed usage patterns, allowing teams to make informed decisions about optimizations.
Security image scanning feature for Vulnerabilities of the images.
Optimise Kubernetes cost with minimal effort
What do you like best about the product?
Cast AI enabled us to significantly reduce our EKS costs, much to the delight of our board of directors. The implementation was remarkably straightforward, and making configuration changes was effortless. Cast AI also adeptly managed our cluster autoscaling needs. Their support is outstanding; the team is always available to help with configuration issues and resolve any Kubernetes challenges we face.
What do you dislike about the product?
I am yet to see any downside of using Cast AI
What problems is the product solving and how is that benefiting you?
We were looking a solution to enable to to scaling our EKS cluster which is when we came across Cast.
Being able to autoscale at the same time as reducing cost has been a huge win for us.
Being able to autoscale at the same time as reducing cost has been a huge win for us.
A Kubernetes Automation Platform That Delivers and Adapts
What do you like best about the product?
CAST AI has proven instrumental in optimizing our Kubernetes operations, leading to significant cost reductions and enhanced efficiency. This platform is not only user-friendly but also effective in managing our Kubernetes clusters, providing exceptional visibility that surpasses other tools we've used. A standout feature is CAST AI's responsiveness; they swiftly address inquiries and provide support via platforms like Slack. Notably, when faced with an unusual situation of IP address shortages, CAST AI quickly identified and resolved a critical feature gap by enabling dynamic subnet additions to our clusters. This adaptability and proactive service significantly alleviated potential operational hurdles.
What do you dislike about the product?
While the benefits of using CAST AI are numerous, new or inexperienced users might find the initial steep learning curve associated with understanding Kubernetes concepts a bit daunting. However, this challenge is mitigated by the comprehensive support and guidance provided by CAST AI's team, making the learning process far more manageable. Beyond this initial phase, I have no significant complaints. CAST AI has proven to be a reliable and proactive partner, consistently addressing our needs and significantly enhancing our operational capabilities.
What problems is the product solving and how is that benefiting you?
CAST AI has been instrumental in addressing two major challenges we faced: efficient cloud resource management and cost optimization. By automating Kubernetes cluster scaling, CAST AI ensured that we use only the necessary resources, which has substantially reduced our cloud expenses. Additionally, their dynamic response to a unique situation involving a shortage of IP addresses—by enabling the addition of new subnets—demonstrated their adaptability and problem-solving capabilities. This not only resolved our immediate issue but also enhanced our trust in their tool's ability to handle unexpected challenges effectively.
The perfect fit to manage your workloads
What do you like best about the product?
- It allowed us to not worry about the size needed for EC2 instances. Without worrying about the resources that the pods used, Cast.ai was able to help us rightsize the instances. Without it this would be a manual job everytime something changed on the workloads, meaning that now we have more time to focus on the overall picture knowing that Cast.ai is there to manage the costs of our infrastructure
- Allowed us to quickly adapt and use different workloads for spot instances for example
- Easy onboard even with the free version that allows to have an immediate knowledge on how much can it be saved
- The possibility to use all the EC2 family types available with small configurations allowed us to quickly save and be ready for future growth and minimize possible cloud failures and lack of resources. For example if the configuration uses all the same family types of EC2 is probable that at some point in time there are not enough machines on the datacenter to provide the same service to multiple clients. Cast.ai allows to select the cheapest instances using filters if needed and always checking if the instances are available, if not, the next cheapest option is used
- Great support from the team with new features appearing constantly. Our feedback is listened :)
- Allowed us to quickly adapt and use different workloads for spot instances for example
- Easy onboard even with the free version that allows to have an immediate knowledge on how much can it be saved
- The possibility to use all the EC2 family types available with small configurations allowed us to quickly save and be ready for future growth and minimize possible cloud failures and lack of resources. For example if the configuration uses all the same family types of EC2 is probable that at some point in time there are not enough machines on the datacenter to provide the same service to multiple clients. Cast.ai allows to select the cheapest instances using filters if needed and always checking if the instances are available, if not, the next cheapest option is used
- Great support from the team with new features appearing constantly. Our feedback is listened :)
What do you dislike about the product?
To be sincere nothing until now happened that I could say that Cast.ai is not a great product
What problems is the product solving and how is that benefiting you?
- It greatly reduced the costs on our side I would say at least 50% depending on the workloads used and every cluster is different.
- With the possibility to predict if SPOT instances would be available we can minimize sudden disruptions on the server, that with a mix of spot instances can help even great cost reductions
- With the possibility to predict if SPOT instances would be available we can minimize sudden disruptions on the server, that with a mix of spot instances can help even great cost reductions
Effortless Cloud Cost Savings with CAST AI
What do you like best about the product?
CAST AI has provided us with instant cost savings and streamlined our Kubernetes management.
Moreover, the team support during the integration process was brilliant. They provided exceptional assistance, ensuring a smooth and successful integration of CAST AI into our existing infrastructure.
I highly recommend CAST AI to any SaaS platform looking to optimize its cloud costs and improve operational efficiency.
Moreover, the team support during the integration process was brilliant. They provided exceptional assistance, ensuring a smooth and successful integration of CAST AI into our existing infrastructure.
I highly recommend CAST AI to any SaaS platform looking to optimize its cloud costs and improve operational efficiency.
What do you dislike about the product?
CastAI Woop had some issues with allocating mem/CPU resources for Java workloads.
What problems is the product solving and how is that benefiting you?
- Cloud Cost Reduction;
- Cluster Autoscaling;
- Efficient Resource Utilization;
- Cost Tracking.
- Cluster Autoscaling;
- Efficient Resource Utilization;
- Cost Tracking.
CAST-AI is the ironsmith of saving
What do you like best about the product?
Keeping track of your CPU is hard, and we started from nodes, watched the drop in price! and now will adopt pods adjustments as well! thank you castAI for your white gloves integration and on going support.
What do you dislike about the product?
I wish they would do my home savings that way as well :)
What problems is the product solving and how is that benefiting you?
They choose the right size of EC2 and pods size, which is a dynamic task which requires logging, monitoring, man in the middle and they do it all.
How cast is able to handle your infra, saving money making and strong and safe infra.
What do you like best about the product?
I like my communication process with the team, and I love to ask for features and support help. They always come to me in less than 5 minutes for calls or messages. The way we delegate to this tool all our infra about gke is impressive we don't worry about the scaling and the recreation of nodes.
Onboarding was super easy because the installation process was very smooth.
Amazing cost savings - we expect to reach savings of around 35% next year due to the fact that we have previously purchased CUDs with Google. They will end next year, and we’re bound to see more savings then.
Onboarding was super easy because the installation process was very smooth.
Amazing cost savings - we expect to reach savings of around 35% next year due to the fact that we have previously purchased CUDs with Google. They will end next year, and we’re bound to see more savings then.
What do you dislike about the product?
Is hard to find something, but I'll point to a specific topic, but this can be a request only for my company; I always miss some information about our infra prices or the total amount of prices related to specific nodes templates or clusters. I think is most related to our cases.
I also think the process to upgrade the tool is no transparent in different cases I found quite of differents UI bugs related to big impacts.
I also think the process to upgrade the tool is no transparent in different cases I found quite of differents UI bugs related to big impacts.
What problems is the product solving and how is that benefiting you?
We have an issue to handle the nodes before using cast, with cast when we need more pods they handle everything.
showing 1 - 10