Cluster Espresso
Home » Cluster Espresso

Cluster Espresso

  • Welcome to our cluster configurator!

    Cluster Espresso will help you to design your cluster. At the end of this wizard, you can request a free budget estimation supplied with our HPC experts device.

    In the next pages you'll be asked to configure:
    1 Cluster type
    2 Rack
    3 Front-end
    4 Computing nodes
    5 Interconnection
    6 Storage
    7 Consolle
    8 Uninterruptible Power Supply (UPS)
    9 Software
    10 Warranty and support service
    Use the Next and Previous buttons to move across the pages.

    Note: you may found explanations about the technical terms used in this wizard inside our glossary

  • 1/10 - Cluster type

    Please, first of all choose the cluster type you need:
    Cluster type
    Let's define the approximate cluster size. This usually requires to choose the number of computing nodes, but Cluster Espresso allows you to specify also other kind of information (available budget, processors number, computing power).
    Number of computing nodes
    Number of processors (cores)
    If you fill these fields too, our HPC experts may optimize the cluster hardware and software configuration and give you useful advices and suggestions:
    Approximate peak performance (GFlops)
    Budget range
    Application field
  • 2/10 - Rack

    Where are you going to put the cluster? The best arrangement is in a rack, inside a conditioned room equipped with a uninterruptible power supply. Please, choose the rack capacity according to the initial cluster size but also taking into account possible future expansions.

    Rack 12 units (60x100x64 cm)
    Rack 25 units (64x80x127 cm)
    Rack 42 units (64x80x202 cm)
    I already have a rack
  • 3/10 - Front-end

    Let's configure the front-end, the main computer acting as an interface between the cluster and its users. It's a good thing to have a RAID1 disk array and a redundant power supply in the front-end.
    No front-end (i.e. this is an upgrade of an existing cluster)
    Front-end
    Chassis
    Number of processors
    Processor type
    RAM (GB)
    Disk type
    Disk size (GB)
    Number of hard-disks
    RAID
    IPMI remote control
    Power supply
  • 4/10 - Computing nodes

    Let's configure the computing nodes. The objective is to obtain the maximum performance for the applications that will be used on the cluster. Please, make your configuration. According to the application field you specified, we could send you our suggestions.

    Computing nodes
    Kind
    Processor type
    RAM (GB)
    Disk type
    Disk size (GB)
    Number of hard-disks
    IPMI remote control
    Power supply
  • 5/10 - Interconnection

    Let's define how to interconnect the cluster nodes. All the nodes are equipped with a couple of Gigabit Ethernet (1Gbit/s) interfaces, you may choose to use this type of networking or another high speed dedicated interconnection. Also in this case, we will take care to suggest you some possible improvements according to the specified application field.

    GigaEthernet Networking
    Number of connections per node
    Dedicated networking
    Kind of network

    Other (please, specify):

  • 6/10 - Storage

    Let's configure the storage. In case you would need a large and a reliable storage unit, in addition to the space available on the front-end, here you may configure a storage.

    Storage unit
    Disk type
    Total capacity (GB)
    RAID
    Connection type
  • 7/10 - Consolle

    Let's choose the consolle, in order to have a monitor, keyboard and mouse connected even just to the front-end. The consolle is mainly used for cluster setup, maintenance and diagnostics. The consolle can be connected to all the cluster nodes through a KVM switch.

    Consolle
    KVM switch
  • 8/10 - Uninterruptible power supply

    Let's choose the uninterruptible power supply. The UPS is important to protect cluster components from sudden voltage changes, to overcome short black-outs and to perform a controlled shutdown in case of long black-outs.

    Insert a value concerning the power in the apposite field (VA or W), if you already own an estimation. Otherwise you may leave this field empty, we will dimension the UPS appropriately.

    Uninterruptible power supply
    Electric power (VA o W)
    Type
  • 9/10 - Software

    Let's configure the software platform. Our company is specialized in delivering ready to use cluster solutions, equipped with all the necessary software.
    Operative system
    Middleware
    More information at Aethia HPC Suite

    Parallel filesystem
    MPI
    Optimized compilers
    Resource manager

    Other software packages, libraries or services required:

  • 10/10 - Warranty and support service

    Please, choose the support service. The basic service includes 2 years of hardware guarantee and 1 year of remote software assistance. You may customize here the support service according to your needs.

    Hardware guarantee
    On-center 2 years warranty (included)
    On-center 3 years warranty
    On-center 3 years warranty NBD (next business day)
    Software assistance
    1 year remote software support (included)
    2 years remote software support
    3 years remote software support

  • Please, choose now how to proceed:
    Offer PDF summary
    By leaving your contact details you'll receive a free estimation with a technical advice from our HPC experts, besides a cluster configuration summary in Pdf format. If you are not interested in a budget estimation, by leaving your email address you'll receive only the cluster configuration summary in Pdf format.
  • The end!

    Cluster configuration has been completed!

    Thank you for using Cluster Espresso.

    By a few days you will receive a free estimation correlated with a technical advice from our HPC experts.

    In order to make this service more and more useful and complete, please leave here a comment or suggestion:


Cluster summary
Racks: 1
Nodes: 0
Processors: 0
Memory: 0 GB
Disk space: 0 TB
Node 42
Node 41
Node 40
Node 39
Node 38
Node 37
Node 36
Node 35
Node 34
Node 33
Node 32
Node 31
Node 30
Node 29
Node 28
Node 27
Node 26
Node 25
Node 24
Node 23
Node 22
Node 21
Node 20
Node 19
Node 18
Node 17
Consolle
Consolle
Switch KVM
Node 16
Node 15
Node 14
Node 13
Node 12
Node 11
Node 10
Node 9
Node 8
Node 7
Node 6
Node 5
Node 4
Node 3
Node 2
Node 1
Front-End
Front-End
Switch 10gb
Switch 10gb
Switch 10gb
Switch 10gb
Switch 10gb
Switch 10gb
Switch FC
Switch FC
Switch FC
Switch FC
Switch FC
Switch FC
Switch IB
Switch IB
Switch IB
Switch 1gb
Switch 1gb
Switch 1gb
Switch 1gb
Storage
U.P.S.