ExoGENI Wiki

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
start [2012/03/28 19:50]
jonmills
start [2021/01/02 06:10] (current)
Line 1: Line 1:
-== Overview of an ExoGENI Rack == +====== Introduction ======
-<rack name=ExoGENI height=42>+
  
-# Head node is at bottom of rack +[[http://www.exogeni.net | ExoGENI]] is based on an extended Infrastructure-as-a-Service (IaaS) cloud model with coordinated provisioning across multiple sites and a high degree of control over intraand inter-site networking functions  Most researchers will use a standard cloud computing stack to instantiate and manage virtual machines 
-1 2 "IBM x3650" xxx-hn.exogeni.net #F5C91B link:doku.php?id=public:headnode:start  Head Node+The sites federate by delegating certain functions for identity management, authorization, and resource management to common coordinator services offered by the ExoGENI federation; ExoGENI in turn delegates some of these functions to the GENI federation and to identity systems operated by participating institutions (Shibboleth/inCommon).
  
-# Worker nodes +This structure enables a network of private ExoGENI IaaS clouds to operate as a hybrid community cloudExoGENI combines this multi-domain cloud structure with rich networking capabilities through direct Layer 2 site connectivity to national circuit backbone fabrics, linkages to other national and international networks, and OpenFlow-enabled dataplanes within each site.
-3 2 "IBM x3650" xxx-w1.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 1 +
-5 2 "IBM x3650" xxx-w2.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 2 +
-7 2 "IBM x3650" xxx-w3.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 3 +
-9 2 "IBM x3650" xxx-w4.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 4 +
-11 2 "IBM x3650" xxx-w5.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 5 +
-13 2 "IBM x3650" xxx-w6.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 6 +
-15 2 "IBM x3650" xxx-w7.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 7 +
-17 2 "IBM x3650" xxx-w8.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 8 +
-20 2 "IBM x3650" xxx-w9.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 9 +
-22 2 "IBM x3650" xxx-w10.site.xo #91B27E link:doku.php?id=public:workernode:start Worker Node 10+
  
-# Keyboard Tray +It provides unified access to these services to enable users to construct virtual network topologies on demand.  The testbed software supports GENI APIs and extended APIs to enable users to create and manage a virtual network as a slice of virtualized resources within the infrastructure.   
-19 1 "Keyboard Pullout" #F55A2A Keyboard Pullout Tray+Basic operations envisioned within ExoGENI include:
  
-# iSCSI Unit +  * Provisioning individual compute resources (virtualized and bare-metal) from rack resources. Users will be able to supply boot images for virtualized instances; bare-metal instances will be limited to a few vetted images.  
-24 2 "IBM DS3512 iSCSI Array" #D8E5F2 link:doku.php?id=public:iscsi:start iSCSI Disk Unit+    * We support Linux (bare-metal and virtual) and Windows (virtual, possibly also bare-metal).  
 +  * Creating, modifying and destroying slices consisting of compute resources belonging to one or more racks, tied together with VLANs provisioned from rack switches and intermediate circuit providers. 
 +  * Create slices with user-driven packet forwarding control via OpenFlow. OpenFlow slices will be restricted to VLANs provisioned within and between the racks. 
 +    * Using OpenFlow, in the longer term we plan to offer an on-ramp feature for allowing external traffic (from campuses or other slices) to transit existing slices (slice-as-a-service). 
 +  * Create slices that combine ExoGENI resources with other GENI resources (e.g. meso-scale OpenFlow and WiMax testbeds; via our switch at the StarLight facility we plan to create experimental L2 topologies that involve international partners). 
 + 
 +Depending on your interests, please use the links to the left to find information you're interested in.
  
-# Data Plane 
-39 1 "BNT 8264" 8264.site.xo #B85356 link:doku.php?id=public:dataswitch:start 48-Port 10Gbps /w 4-Port 40Gbps Data Plane Switch 
-# Mgt Switch 
-41 1 "BNT 8052" 8052.site.xo #C072CD link:doku.php?id=public:mgtswitch:start 48-Port 1Gbps Management Switch 
-# Blank 
-42 1 "" #EEEEEE 
-</rack> 
Navigation
Print/export
QR Code
QR Code start (generated for current page)