This site uses cookies to enhance your web site experience. By continuing to browse or use this site, you are agreeing to our use of cookies. See our privacy policy here.

Engineered for tomorrow’s AI

BoosTrak C6000 series is a high-performance, scalable computing platform, engineered to handle the demanding workloads of modern AI applications.
Equipped with multi-core CPUs, multiple high-end GPUs or accelerators, and terabytes to petabytes of high-speed NVMe flash storage, it delivers massive parallel processing power for training, fine-tuning, and inference across large language models, computer vision systems, and multimodal AI pipelines.

Advance Platform for Advance AI

Built on the latest Intel® Xeon® 6 platform, with high-speed DDR5 memory and PCIe 5.0 internal bus fabric, the BoosTrak C series provides top-of the line compute and storage server platform for advance AI applications.

Be it AI training, inference, Agentic AI, RAG or HPC clusters, BoosTrak C series present a reliable and stable platform.

Balanced CPU, DDR and GPU VRAM Ratio

The BoosTrak C6300 and C6400 platforms provide critical balance between GPU and CPU for Inference and Agentic AI. A Unique 3U and 4U enclosure that accommodate up to 4 large GPUs, are backed by single or dual CPU motherboard.

Storage and Networking Flexibility

  • Unique design for storage and network.
  • BoosTrak GPU Compute servers achieves a unique feet, that houses hot data from lighting speed PCIe 5.0 NVMe SSDs and low speed archive storage, all under single enclosure and server.
  • Multiple networking options that reach up to 400GbE smoothly integrates the compute server into the cluster, rack and rest of the datacenter. 

A Dual Node Storage Server of Its Own Kind

A purpose built dual/single node storage server BoosTrak C6200 is engineered as a high-speed data foundation for modern AI infrastructure.

Fully independent server nodes consolidated into a single, power-efficient chassis, combining PCIe 5.0 NVMe flash, high-bandwidth networking and SAS-4 JBOD expansion to feed GPU clusters with the sustained throughput that training and inference pipelines demand.

One Server to Cover Them All

A complete server platform that is right for any storage environment or architecture. Be it the high performing parallel file systems, or large capacity data lakes, distributed system, or virtual environment, the BoosTrak C6000 products are best suited solution.

Stringently tested in the lab under various solutions and environment, such as, XFS, ZFS, Lustre, DAOS, EOS, BeeGFS, WEKA, VAST, VMware, Proxmox, OpenStack, Azure, S3, and many more, the BoosTrak and VTrak products offers the compatibility, reliability and performance that the advance AI applications, and critical data centers demand.

PROMISE Advantages

Massive Hyper-scale Data centers to edge AI applications, Promise BoosTrak series products provide complete coverage.

Accelerated data pipeline with NVMeBoost 2.0

Promise’s NVMeBoost 2.0 accelerates the data transfer between NVMe SSD and data starving GPUs. Using tools, technologies, the NVMeBoost 2.0 makes it intuitive and easy to manage the complex data pipeline. The AI model and algorithm of NVMeBoost 2.0 optimize per watt performance.

Sustainable AI with GreenBoost 2.0

The BoosTrak products are design and integrated with Promise’s ground breaking technologies, to optimise and accelerate the AI data centers. AI Model based prediction algorithm of GreenBoost 2.0 not only does the deep energy saving, but also reduce cooling cost and lower the heat generation.

PROMISE BoosTrak series gives a single stop solution for planning, deployment and operation of AI workflow, from datacenter to edge.

Target Applications

The BoosTrak Family is designed to support data-intensive workloads across AI, HPC, and high-performance enterprise environments.

AI Model Training

Supports high-throughput data pipelines required for large-scale AI model training, enabling rapid access to massive datasets and accelerating machine learning workflows.

High-Performance Computing (HPC)

Delivers ultra-low latency and high parallel I/O performance for demanding HPC environments, including scientific simulations, engineering analysis, and research workloads.

Real-Time Data Analytics

Enables real-time data processing and analytics by providing fast and consistent access to large datasets for data-driven decision-making.

High-Performance Databases

Accelerates transactional and analytical databases that require high IOPS and low latency to maintain performance under heavy workloads.

Media Production and Rendering

Handles large media datasets for video rendering, visual effects, and content production workflows that demand high-throughput storage performance.

Models

  • Features
    • Dual Node Storage Server
    • PCIe5.0 NVMe SSD
    • Flexible Ultra High Bandwidth
    • SAS-4 JBOD Expansion
    • Intel® Xeon® 600 series processor
  • Specifications

    Specifications

    Form Factor
    2U 19" Rackmount
    Processor
    Intel® Xeon® 600 processor per Node (Default processor Intel® Xeon® 634)
    Memory
    Single Node: 8-Channel, 8 x DIMM slots
    Dual Node: 8-Channel, 8 x DIMM slots per Node
    Internal Storage
    Single Node: 2 x M.2 (2280/22110), PCIe Gen4 x4
    Dual Node: 2 x M.2 (2280/22110), PCIe Gen4 x4 per Node
    Number of Bays
    Single Node: 24 bays
    Dual Node: 12 bays per Node. Total 24 bays
    Drive Support
    U.2 / U.3 NVMe SSD
    PCIe Slots per Node
    Slot 1: Standard PCIe Gen5.0 x16 (x8 when Slot 2 is in use)
    Slot 2: Standard PCIe Gen5.0 x8
    Slot 3: Standard Gen4.0 x4
    Supported GPU1
    Single Node: Up to 1 x Single Slot HLHH GPU
    Dual Node: Up to 1 x Single Slot HLHH GPU per Node
    Host Interface Options per Node
    1. Quad port 25GbE SFP28, 2. Dual port 100 QSFP112
    USB Ports
    2 x USB 3.2 Gen1 (rear) per Node
    Storage Expansion
    2 x SFF-8674 per Node
    Display Port
    1 x VGA port per Node
    OS Compatibility
    Windows Server, Debian, Ubuntu, Red Hat Enterprise, Rocky Linux, AlmaLinux
    Platforms and Architecture
    XFS, ZFS, Lustre, DAOS, EOS, BeeGFS, WEKA, VAST, VMware, Proxmox, OpenStack, Azure, S3
    Management Interfaces
    1 x RJ45 1GbE per Node, 1x RJ11 UART per Node, 1 x RJ45 1GbE per Node (over BMC HBA)
    Management Tool2
    PROMISE WebPAM 5.0, CLI, Redfish Management APIs, 3rd party management tools
    Promise Features3
    OPAS, NVMeBoost, GreenBoost
    Power Supply
    Dual 80 Plus Platinum 1600W CRPS Redundant Power Supply
    AC Input
    AC 100-240V, 60~50Hz
    Dimensions (H x W x D)
    88.4 x 484.2 x 773.5 mm
    Temperature Range
    Operational: 5° to 35°C / Non-Operational: -40° to 60°C
    Field Replaceable Unit (FRU)
    Drive Bay Modules, Power Supply Unit Modules, Server Node Module
    Warranty
    3-years standard warranty, optional extended warranty, and on-site parts replacement program (PROMISE ServicePlus) available.
    Support
    24/7 PROMISE eSupport Portal (support.promise.com), optional 24/7 phone and email support available.

    Note:

    1. GPU support is conditional to the storage and network interface options.
    2. Promise Management tools are subject to the selected OS and customization.
    3. Promise Optimization Features are subject to OS, Application and Use Case.

    Please note that product specifications may change without prior notice. In case of discrepancy, the English datasheet takes precedence.

  • Resources

    Documents

Copyright ©2026 PROMISE Technology Inc. All rights reserved. / Terms Of Use / Privacy Policy / Contact Us