US2023362063A1PendingUtilityA1

Discovery of physical network architecture

Assignee: VMWARE INCPriority: Mar 29, 2022Filed: Jul 19, 2023Published: Nov 9, 2023
Est. expiryMar 29, 2042(~15.6 yrs left)· nominal 20-yr term from priority
H04L 41/22H04L 41/0803H04L 41/12H04L 41/0893H04L 41/122
57
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

Some embodiments provide a method for identifying network architecture in a datacenter. The method identifies connection information for multiple network elements. The connection information for each network element specifies other network elements to which the network element connects. The method assigns a set of the network elements to a set of two or more layers based on the connection information for the network element. The method uses the connection information to identify at least one group of network elements organized into a particular network architecture. The identified groups are for use in displaying user interface visualizations of the network elements of the datacenter.

Claims

exact text as granted — not AI-modified
1 - 25 . (canceled) 
     
     
         26 . A method for identifying and displaying network architecture in a datacenter, the method comprising:
 using connection information for a plurality of network elements of a datacenter underlay network to assign each of a set of the network elements to one layer of a set of two or more layers of the datacenter underlay network, the connection information for each network element specifying other network elements to which the network element connects;   identifying one or more groups of network elements organized into a leaf-spine network architecture based on the assigned layers and the connection information; and   using at least one of the identified groups to display a user interface visualization of a data message path from a first data compute node executing on a first host computer in the datacenter, through a subset of the network elements in the leaf-spine architecture of the identified group, and to a second data compute node executing on a second host computer in the datacenter, wherein the visualization of the data message path comprises visualizations of both (i) a set of logical network elements implemented by the physical host computers and (ii) the subset of physical network elements in the leaf-spine architecture of the identified group.   
     
     
         27 . The method of  claim 26 , wherein the datacenter underlay network carries data messages between a plurality of host computers for a plurality of different logical overlay networks implemented in the datacenter. 
     
     
         28 . The method of  claim 27 , wherein the set of logical network elements belong to a particular one of the logical overlay networks and are implemented by a set of software forwarding elements executing on a set of the host computers, including the first and second host computers. 
     
     
         29 . The method of  claim 26 , wherein:
 each group of network elements comprises a set of leaf-layer network elements and a set of spine-layer network elements; and   the subset of network elements comprises at least two leaf-layer network elements and at least one spine-layer network element.   
     
     
         30 . The method of  claim 29 , wherein (i) the user interface displays visualizations of all of the network elements in the group of network elements, (ii) the visualizations of the network elements indicate the set of leaf-layer network elements and the set of spine-layer network elements, and (iii) the visualization of the data message path indicates the subset of network elements through which the data message passes. 
     
     
         31 . The method of  claim 26 , wherein the connection information is learned based on monitoring of traffic in the datacenter network and does not require receiving any configuration data from the network elements. 
     
     
         32 . The method of  claim 26 , wherein using the connection information to assign the network elements to layers comprises:
 identifying a first set of network elements in a first layer closest to sources and destinations of data traffic in the network; and   identifying a second set of network elements in a second layer that provide connections between the network elements in the first layer,   wherein (i) each first-layer network element directly connects to at least one second-layer network element and does not connect to any other first-layer network elements in the first layer; and (ii) each second-layer network element directly connects to at least one first-layer network element and does not connect to any other second-layer network elements.   
     
     
         33 . The method of  claim 26 , wherein identifying one or more groups of network elements organized into a leaf-spine network architecture comprises:
 defining a set of matrices representing connections between sets of network elements assigned to a first layer and sets of network elements assigned to a second layer; and   using the set of matrices to identify groups of network elements in which each network element in the first layer connects to each network element in the second layer.   
     
     
         34 . The method of  claim 33 , wherein:
 the set of matrices comprises (i) a first matrix representing connections between a first set of network elements assigned to a first layer and a second set of network elements assigned to a second layer and (ii) a second matrix representing connections between a third set of network elements assigned to a first layer and a fourth set of network elements assigned to a second layer; and   using the set of matrices to identify groups of network elements comprises:
 using the first matrix to identify that each of the network elements in the first set of first-layer network elements connects to each of the network elements in the second set of second-layer network elements; 
 using the second matrix to identify that each of the network elements in the third set of first-layer network elements connects to each of the network elements in the fourth set of second-layer network elements; and 
 identifying the first set of first-layer network elements and the second set of second-layer network elements as belonging to a first group of network elements organized into a leaf-spine architecture and the third set of first-layer network elements and the fourth set of second-layer network elements as belonging to a second group of network elements organized into a leaf-spine architecture. 
   
     
     
         35 . The method of  claim 34 , wherein the set of matrices is a first set of matrices, wherein identifying one or more groups of network elements further comprises:
 defining a second set of matrices representing connections between sets of network elements assigned to the second layer and sets of network elements assigned to a third layer, the second set of matrices comprising a third matrix representing connections between a fifth set of second-layer network elements and a sixth set of network elements assigned to the third layer, the fifth set of second-layer network elements comprising both the second set of second-layer network elements and the fourth set of second-layer network elements; and   using the second set of matrices to identify groups of network elements in which each network element in the second layer connects to each network element in the third layer,   wherein using the second set of matrices comprises:
 using the third matrix to identify that each of the network elements in the fifth set of second-layer network elements connects to each of the network elements in the sixth set of third-layer network elements; and 
 identifying the first set of first-layer network elements, third set of first-layer network elements, fifth set of second-layer network elements, and sixth set of third-layer network elements as belonging to a three-layer group of network elements organized into a leaf-spine-superspine architecture. 
   
     
     
         36 . A non-transitory machine-readable medium storing a program which when executed by at least one processing unit identifies and displays network architecture in a datacenter, the program comprising sets of instructions for:
 using connection information for a plurality of network elements of a datacenter underlay network to assign each of a set of the network elements to one layer of a set of two or more layers of the datacenter underlay network, the connection information for each network element specifying other network elements to which the network element connects;   identifying one or more groups of network elements organized into a leaf-spine network architecture based on the assigned layers and the connection information; and   using at least one of the identified groups to display a user interface visualization of a data message path from a first data compute node executing on a first host computer in the datacenter, through a subset of the network elements in the leaf-spine architecture of the identified group, and to a second data compute node executing on a second host computer in the datacenter, wherein the visualization of the data message path comprises visualizations of both (i) a set of logical network elements implemented by the physical host computers and (ii) the subset of physical network elements in the leaf-spine architecture of the identified group.   
     
     
         37 . The non-transitory machine-readable medium of  claim 36 , wherein the datacenter underlay network carries data messages between a plurality of host computers for a plurality of different logical overlay networks implemented in the datacenter. 
     
     
         38 . The non-transitory machine-readable medium of  claim 37 , wherein the set of logical network elements belong to a particular one of the logical overlay networks and are implemented by a set of software forwarding elements executing on a set of the host computers, including the first and second host computers. 
     
     
         39 . The non-transitory machine-readable medium of  claim 36 , wherein:
 each group of network elements comprises a set of leaf-layer network elements and a set of spine-layer network elements; and   the subset of network elements comprises at least two leaf-layer network elements and at least one spine-layer network element.   
     
     
         40 . The non-transitory machine-readable medium of  claim 39 , wherein (i) the user interface displays visualizations of all of the network elements in the group of network elements, (ii) the visualizations of the network elements indicate the set of leaf-layer network elements and the set of spine-layer network elements, and (iii) the visualization of the data message path indicates the subset of network elements through which the data message passes. 
     
     
         41 . The non-transitory machine-readable medium of  claim 36 , wherein the connection information is learned based on monitoring of traffic in the datacenter network and does not require receiving any configuration data from the network elements. 
     
     
         42 . The non-transitory machine-readable medium of  claim 36 , wherein the set of instructions for using the connection information to assign the network elements to layers comprises sets of instructions for:
 identifying a first set of network elements in a first layer closest to sources and destinations of data traffic in the network; and   identifying a second set of network elements in a second layer that provide connections between the network elements in the first layer,   wherein (i) each first-layer network element directly connects to at least one second-layer network element and does not connect to any other first-layer network elements in the first layer; and (ii) each second-layer network element directly connects to at least one first-layer network element and does not connect to any other second-layer network elements.   
     
     
         43 . The method of  claim 36 , wherein the set of instructions for identifying one or more groups of network elements organized into a leaf-spine network architecture comprises sets of instructions for:
 defining a set of matrices representing connections between sets of network elements assigned to a first layer and sets of network elements assigned to a second layer; and   using the set of matrices to identify groups of network elements in which each network element in the first layer connects to each network element in the second layer.   
     
     
         44 . The non-transitory machine-readable medium of  claim 43 , wherein:
 the set of matrices comprises (i) a first matrix representing connections between a first set of network elements assigned to a first layer and a second set of network elements assigned to a second layer and (ii) a second matrix representing connections between a third set of network elements assigned to a first layer and a fourth set of network elements assigned to a second layer; and   the set of instructions for using the set of matrices to identify groups of network elements comprises sets of instructions for:
 using the first matrix to identify that each of the network elements in the first set of first-layer network elements connects to each of the network elements in the second set of second-layer network elements; 
 using the second matrix to identify that each of the network elements in the third set of first-layer network elements connects to each of the network elements in the fourth set of second-layer network elements; and 
 identifying the first set of first-layer network elements and the second set of second-layer network elements as belonging to a first group of network elements organized into a leaf-spine architecture and the third set of first-layer network elements and the fourth set of second-layer network elements as belonging to a second group of network elements organized into a leaf-spine architecture. 
   
     
     
         45 . The non-transitory machine-readable medium of  claim 44 , wherein the set of matrices is a first set of matrices, wherein the set of instructions for identifying one or more groups of network elements further comprises sets of instructions for:
 defining a second set of matrices representing connections between sets of network elements assigned to the second layer and sets of network elements assigned to a third layer, the second set of matrices comprising a third matrix representing connections between a fifth set of second-layer network elements and a sixth set of network elements assigned to the third layer, the fifth set of second-layer network elements comprising both the second set of second-layer network elements and the fourth set of second-layer network elements; and   using the second set of matrices to identify groups of network elements in which each network element in the second layer connects to each network element in the third layer,   wherein the set of instructions for using the second set of matrices comprises sets of instructions for:
 using the third matrix to identify that each of the network elements in the fifth set of second-layer network elements connects to each of the network elements in the sixth set of third-layer network elements; and 
 identifying the first set of first-layer network elements, third set of first-layer network elements, fifth set of second-layer network elements, and sixth set of third-layer network elements as belonging to a three-layer group of network elements organized into a leaf-spine-superspine architecture.

Join the waitlist — get patent alerts

Track US2023362063A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.