AI Training Workloads Network Deployment

₹1000000-2500000 INR

Closed

Posted 6 days ago

₹1000000-2500000 INR

Paid on delivery

Technical Summary Architecting, deploying, and validating a non-blocking 2-tier Spine-Leaf Clos fabric designed for distributed AI training workloads. The network utilizes BGP-EVPN with VXLAN overlay alongside end-to-end RoCEv2 lossless Ethernet parameters (PFC, ECN, DCBX) to ensure zero packet loss and sub-2 microsecond latency across NVIDIA GPU clusters. Technical Details Topology & Core Fabric - Architecture: Non-blocking 2-tier Spine-Leaf Clos topology - Underlay Routing: eBGP with IPv4 loopback reachability - Overlay Control Plane: BGP-EVPN (Type-2 MAC/IP & Type-5 IP Prefix routes) - Data Plane Encapsulation: VXLAN - Global MTU: 9216 bytes (Jumbo Frames end-to-end) QoS & Lossless Ethernet Mechanics (RoCEv2) - Class of Service Mapping: CoS 3 reserved for RDMA traffic (DSCP 26 / AF31) - Priority-Based Flow Control (PFC): IEEE 802.1Qbb enforced on CoS 3 - Explicit Congestion Notification (ECN): RFC 3168 RED marking - K_min (Min Threshold): 150 KB - K_max (Max Threshold): 1500 KB - PFC Headroom Trigger: 3000 KB - Parameter Negotiation: DCBX (IEEE 802.1Qaz) Hardware & Platform Specifications - Compute & Offload: NVIDIA A100/H100 GPUs, NVIDIA ConnectX-6/7 SmartNICs (GPUDirect RDMA / GDR) - Switching Infrastructure: Arista EOS, Cisco NX-OS, and disaggregated NOS platforms - Testing Infrastructure: Ixia IxNetwork traffic generators (line-rate 100G/200G stress testing)

AI (Artificial Intelligence) HW/SW

AI Workflow Automation

Artificial Neural Network

Project ID: 40744441

About the project

4 proposals

Remote project

Active 6 days ago

Looking to make some money?

Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

Get paid for your work

Outline your proposal

It's free to sign up and bid on jobs

4 freelancers are bidding on average ₹1,343,750 INR for this job

@rajeshrolen

Hi, At first glance, this looks straightforward but there’s usually one part that causes issues later. I’ve handled similar work before and can help you avoid that. Regards, Rajesh

₹1,000,000 INR in 7 days

5.0

(1 review)

2.0

@artechvisionn

Hey! I can handle architecting and deploying your 2-tier Spine-Leaf Clos fabric network easily. I've designed and validated similar AI training workload networks about 12 times recently, ensuring non-blocking performance and low latency for heavy AI model data flows. How I'd approach this: 1. Assess your current hardware and AI training workflow requirements. 2. Design the network topology with optimal spine and leaf switches to handle AI data traffic without bottlenecks. 3. Deploy and configure the network, including routing and redundancy, to guarantee high availability. 4. Validate the setup through stress tests simulating your AI training workloads to confirm throughput and stability. Could you share details about your existing network hardware and AI training data volumes? That will help tailor the design precisely. I understand the critical role of network performance in AI model training and can ensure your deployment supports seamless data movement and integration. Aakash P.S. Happy to discuss your specific AI training needs and network goals anytime. Just send a message!

₹1,000,000 INR in 14 days