Vertical Mobility: Inference from MVP to Trillion-Parameter Workloads — Sitanshu Gupta, CoreWeave

0:13 Good afternoon everybody. I'm Satan Shu from Kov going to be talking about vertical mobility. It's a quite a fancy topic the title that we came up with but basically going to be talking about the inference platform that we have at Corv that we are building to serve small to big models and various different types of workloads. a quick intro about me. I joined core just about four months back. leading all of inference over there and before this I was managing everything at AWS Anapuna labs for training and before that inference and training at SANOVA. So quite a bit of experience in in this particular space.

0:55 what I will the way I'll be taking you through is explaining to you the consumption models that we have and from that how we have derived what the platform should look like so that we do not need to keep changing the platform and we keep making enhancements in the platform that we have for serving inference and how and why the why performance plays such an important role over there. I think a little bit of this might be common with the previous topic that was discussed over here.

1:21 consumption models. So we have at large two biggest consumption models. One is a serverless which is where the customers can come in, consumers can come in do not need to worry about managing the hardware themselves. Do not need to worry about managing the clusters orchestration anything at all. There's API, there's UI, you come in, you pay per token and you get your model served. biggest thing over here is that the type of models that we serve in the catalog that is the breadth of models that the customer will be able to go through. I'll talk about dedicated and then I'll come back to serverless because there is one thing unique on the serverless side. Dedicated inference service that we provide is more for customers who want to know exactly what hardware they are going to be using and running on. but the model deployment also depends on them.

2:15 So they use our service, they use our orchestration layers. but the model deployment depends on them. Model performance also depends on them as long as we provide in the platform the capability and the knobs to serve those features. Coming back to serverless, one of the interesting pieces over here is typically serverless models are very are noisy neighbor problems where if let's say everyone is banging on the exact same model then you might be timing out quite a bit depending on how much capacity I have behind it. So another feature that we have on the serverless side is what we calling provisioned throughput. So as a customer, if you know your traffic profile and if you can let us know about that, we can carve it out specifically for you behind the scenes. You still do not need to worry about what hardware it is exactly running on as long as your throughput, your SLAs's are maintained.

3:08 So that is another one on the serverless side and that is still charged by by the token but you know that you're not running into the noise enabled problem over there. let me take a quick stab at few different types of workloads workload shapes that we have that we are seeing and the ratio between these is like continuously changing though agentic is like really high up there. agentic and chat kind of very similar. Super high on the input sequence lengths, very low on the output sequence lengths typically. But the biggest difference between agentic and chat being the fact that the multi-turns in agentic are super low latency versus in chats because when you get the response as a user, you have to read the answer and then you respond to it. So there are there are differences over there and that big difference ultimately converts into something related to the KV cache management but these are these two are both real time and another real-time workload is your voice and videos which are study streaming and super latency sensitive on the agent agent and chat side largely the requirements are from throughput point of view not so much from latency but real-time voice and videos are absolutely totally latency sensitive.

4:26 Coming to batch, batch is where the SLAs's are like super loose. They run into like way like seconds and minutes. Sometimes for some customers actually even in hours. They're like I'll just throw give me 10 to 12 hours of workload capability and I'll throw whatever I can process it whenever you can. these batch workloads the way they come into the picture over here in deciding sorry being the requirement for some of the design choices that we make in the stack. Imagine these four different types of workload shapes in the time dimension you have to play the game of tetris on how you can fit it in to utilize the underlying infrastructure the most.

5:12 I'll give a high level on how our stack is shaped right now. and I'll walk you through a bit of a request flow over here. So for both serverless and dedicated, if you look at the right hand side of the screen, you'll see that on the platform side we you'll go to the control plane to to have your authorizations, your rate limitings and your usage being tracked etc. so that we can be built accordingly and observability so that we can make sure that we are not violating the SLAs's that have been signed right underlying on the platform I've shown at a super high level that we have these different inference engines VLMs SG langs and tensorl but there are quite there's quite a few quite a bit of detail over here that I'll touch upon and underlying that what I'm trying to show over here in green is various different pieces of hardware. So it's not that so the platform needs to be capable enough of share of having the workload getting distributed across various different generations of these GPUs specifically Nvidia GPUs that we use right so let's take up a let's take a few examples over here let's say the request originates from the client side through apps or notebooks any of those or through the agents right it hits the gateway once it hits the gateway then like I mentioned on the control plane goes through authentication etc etc etc but then comes either the serverless or dedicated so in the case of serverless it'll be paper token so that the token usage would be monitored over here not the exact tokens but just the token usage because we maintain ZDR zero data retention policies it is multi depending on the multi-tenency or the provisioned if it is provisioned then we know underlying for the router it needs to go in and target the the explicit deployments for the provision throughput customers. For the multi-end customers, there are separate deployments.

7:17 Router over here specifically the the router is very important since the router is responsible for making KV cache aware routing choices. Why is it important? Because like I mentioned when we were discussing the workload profiles the agentic use cases are typically super heavy on the input sequence lengths and bulk of the input sequence length about 80 to 90% depending on which company it is depending on the customers 80 to 90% of it is the same for various different requests. So there is no point in going in and recomputing the prefill or redoing the prefill for that prefill is supercomputebound very expensive that's why as much as you can hit the cache more you can save which is why if you look at the token pricing anywhere there's a specific price for input tokens and there's a way cheaper price for the cash input tokens so caching becomes like really important over here underlying The underlying how you want to split the hardware is totally dependent on the choice in the platform and we provide the capability to do either either do a pre-filled decode disagregation if the use case desires it or do not do it because prefilled decode disagregation is not cheap for every type of use case.

8:42 let's take another request flow over here. Let's see if when it was a dedicated customer then what will happen? dedicated customer again will go through the gateways that have been set up for them with proper isolations. billing is not based on tokens. Billing is based on usage of per GPU per hour. it's a private gateway so that there is noisy enable problem. No one else can get in. same router logic over here so that if there are cache heavy if there are requests which are very similar then it's it hits the cache most and depending on the deployment that the customer makes in their on their dedicated GPUs they can decide if they want to do prefill decode disagregation or not they can decide which which engine to use VLM or S lang or tenslm and given the bulk of capacity that the customer has reserved they can decide if they want to have just one deployment with the ability to scale through the whole cluster or they want to have multiple different models, multiple different deployments.

9:49 one thing that I do want to mention about the router over here is the fact that heterogeneous capacity across different zones and regions is supported. It is a little it's quite a bit of a hard problem to load balance across that. So the priority order that we typically take is first KV cache locality and then the least loaded fallback. that's that another request flow that I want to go over here which might be a little hard to see from the diagram is I want to take the batch workflow. For the batch workflow what we would actually do is the underlying capacity that the customer has let's say the same dedicated inference customer during US daytime they're running their real-time workloads and from evening to night they want to run batch workloads the same capacity after time can be scheduled to run the batch workloads. So we provide the capability in the API to tell when to scale up and when to scale down and as per schedule if we can if they tell us that we have to scale down we will scale down and open it up for batch processing through the night.

11:05 I think I've spoken quite a bit about optimizations on the KV cache side but I do want to repeat a little bit because this is one of the most interesting pieces. it if we can hit on the cache more you can you will avoid the cost of prefill which is the most expensive piece over here. reusing the cache across multiple different turns in your agentic workloads between turns also there is lot of similar prefill that comes in in the input sequence length.

11:38 think about the chat workloads which is where offloading AV cache also becomes extremely important because with cache with with the chat workloads we have a lot of latency between different between multiple turns that we as users put in but if we completely evict whatever we had in our particular conversation then the next time we ask a question in that same chat it's going to take a little bit longer. so instead of actually completely evicting and redoing the prefill again what the the techniques being used are maybe using u we we are using our own but externally we know about LM cache and moon cakes.

12:17 what we do is we will offload the KV cache to a high bandwidth storage so that we can store a lot of these prefills such that whenever the accompanying request comes for that particular conversation it can be loaded in right away into the HPM. on the performance liver I would I just want to mention a few performance livers. We've discussed the PD disagulative decoding are others and how to carefully choose the parallelization degrees and the strategies that is actually very important. two of the biggest livers that we've been working with are quantization to NVFP4 and spec.

13:01 we do provide capability where if the customer has their data set and they want us to train in speculators for their data sets for better acceptance lens which will ultimately make the output throughput significantly higher. we do have that as well. So but that happens async. We we get the data async we train the speculators async and then we deploy the speculators into the customer deployments if that's what they wanted. you see three screenshots over here. I have posted them from the last one month one month's worth of work that some of us in my team have done. You can see we came quickly on top of the leaderboard on Kimmy 2.6 2.7 and those are those are from artificial analysis and going back to the session before this can we trust that that's why for GLM I have the results from open router. So artificial analysis when they run benchmarks they're running very specific workloads open router is actual user traffic and you can see on the open router side weights and biases so the branding is different but weights and bias is basically k we bought weights and biases about a year back you can see the speed over here that we have from our deployment is pretty close to what fireworks is providing us fireworks fast right but underlying techniques that we are using is what I want to emphasize the most over here for for performance optimization that becomes critical because ultimately what you want to serve to the customer what we want to serve to the customer is price performance benefit quick recap single platform is what I've been trying to emphasize is what I've tried to show two different consumption models serverless and dedicated for customers and within serverless I describe two different consumption models as well pay as you go and provision vision throughput if you care about that and ultimately compounding the gains through performance optimizations in the stack.

15:02 That's all. Thank you folks. >>