IL280293A - A method for implementing a consistent hashing in a communication network - Google Patents

A method for implementing a consistent hashing in a communication network

Info

Publication number
IL280293A
IL280293A IL280293A IL28029321A IL280293A IL 280293 A IL280293 A IL 280293A IL 280293 A IL280293 A IL 280293A IL 28029321 A IL28029321 A IL 28029321A IL 280293 A IL280293 A IL 280293A
Authority
IL
Israel
Prior art keywords
consistent hashing
virtual identification
packet
processing units
communication network
Prior art date
Application number
IL280293A
Other languages
Hebrew (he)
Inventor
Krayden Amir
LEV Yuval
SANDLER Evgeny
Zilberman Alexander
Original Assignee
Drivenets Ltd
At& T Services Inc
Krayden Amir
LEV Yuval
SANDLER Evgeny
Zilberman Alexander
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Drivenets Ltd, At& T Services Inc, Krayden Amir, LEV Yuval, SANDLER Evgeny, Zilberman Alexander filed Critical Drivenets Ltd
Priority to IL280293A priority Critical patent/IL280293A/en
Publication of IL280293A publication Critical patent/IL280293A/en

Links

Description

A METHOD FOR IMPLEMENTING A CONSISTENT HASHING IN A COMMUNICATION NETWORK TECHNICAL FIELD The present disclosure generally relates to the field of communication systems. More particularly, the present disclosure relates to systems implementing infrastructure for multiservice routing, virtual network function virtualization and software- defined networking.
GLOSSARY ECMP- Equal Cost Multi-Path. EPC- Evolved Packet Core. FRR- Fast Re-Route. LPM- Longest Prefix Match. NAT- Network Address Translation. NCP- Network Cloud Packet Forwarder. NPU- Network Processing Unit. TCAM- Ternary Content-Addressable Memory. WCMP- Weighted Cost Multi-Path. VRF– Virtual Routing and Forwarding.
BACKGROUND High-scale stateful load balancing is an important part of many network functions and an active field of academic and practical research. The use of load balancing is part of many different layers in networking functionalities, including but not limited to: Link-aggregation traffic distribution, ECMP / WCMP (Equal Cost Multi-Path, Weighted Cost Multi-Path), L4-LSession Load Balancing, Distributed Data-base implementation, Mobility EPC (Evolved Packet Core) functions, NAT (Network Address Translations) functions, and the like.Classical non stateful (Link-aggregation, ECMP / ECMP) or stateful (L4-L7 Session Load Balancing, etc.) are all revolving around the way to choose the destination or eventual server / micro-service which should handle a traffic item. The way in which the existing destinations are obtained, managed, accounted for or being classified are beyond the scope of this disclosure, so are the exact ways by which load parameters / abilities of these destinations are conveyed to the load balancing device / devices (which can be based on out-of-band communication or in­band communication using special streams or piggi-bagging existing ones, where problems are more complicated in case of a distributed load-balancing architecture).
SUMMARY The disclosure may be summarized by referring to the appended claims.It is an object of the present disclosure to provide a novel and improved method that allows affecting changes in a communication network comprising a plurality of network processing units and/or clusters thereof, while adding/removing/overcoming a failure of a network processing unit.It is another object of the present disclosure to provide a novel method that enables obtaining an improved load balancing performance.It is another object of the present disclosure to provide a novel method that enables obtaining an improved scaling up of existing networks that comprise a vast number of network processing units.It is yet another object of the present disclosure to provide a novel and improved method for utilizing different load-balancing groups in conveying communications in a communication network.Other objects of the present disclosure will become apparent from the following description.
According to a first embodiment of the present disclosure there is provided a method for use in a communication network, wherein the method comprises the steps of:(i) providing a plurality of network processing units (NPU's) comprised in the communication network;(ii) establishing a replication of at least one of that plurality of network processing units ;(iii) virtually arranging the plurality of network processing units and the at least one replication, in a ring configuration; (iv) associating a unique primary virtual identification and a corresponding unique backup virtual identification with each active and available entity selected from among the plurality of network processing units and the at least one replication;(v) establishing a list of hash values, each associated with the primary virtual identification or with the backup virtual identification of a corresponding active and available entity;(vi) implementing a ring consistent hashing algorithm for carrying out a search resolution for a consistent hashing (e.g., by using an arbitrary key K); and(vii) in a case of a change (i.e., adding/removing/failure) in at least one of the active and available entities having a respective unique primary virtual identification, using the corresponding backup virtual identification to maintain the ring continuity.
By yet another embodiment, the search resolution for a consistent hashing is conducted based on searching for a packet destination ID using a key built from said packet’s meta-data. Optionally, the packet’s meta-data is derived from the packet's networking headers.According to still another embodiment, the search resolution for a consistent hashing is done by retrieving a respective destination ID while using a Ternary Content- Addressable Memory (TCAM) algorithm.
In accordance with another embodiment, the method provided further comprises a step of composing a route space of the networking subnets and/or prefixes, and wherein the search function is based on a Longest Prefix Match (LPM) algorithm.According to yet another embodiment, the same search function (LPM) is preserved, and the space is built by using possible virtual identifications as prefixes.By still another embodiment of the present disclosure, the method provided further comprises a step of utilizing different load-balancing groups as VRF (Virtual Routing and Forwarding) tables, which in turn are used as a parameter while carrying out the LPM algorithm search, thereby generating a dedicated consistent hashing space for use as a route space for conveying traffic in the communication network.
BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawing, which is incorporated herein and constitute a part of this specification, illustrates an embodiment of the disclosure and, together with the description, serve to explain the principles of the embodiments disclosed herein. FIG. 1 exemplifies a ring virtual configuration of software consistent hashing space, construed in accordance with an embodiment of the present invention; FIG. 2 exemplifies a search resolution for a software consistent hashing using an arbitrary key, for the network exemplified in FIG.1; FIG. 3 exemplifies a search resolution for a software consistent hashing using an arbitrary key after a node failure; FIG. 4 exemplifies a ring virtual configuration of software consistent hashing space for a network comprising NPUs; FIG. 5 exemplifies a search resolution for a software consistent hashing using an arbitrary key, for the NPUs network exemplified in FIG.4; FIG. 6 exemplifies a search resolution for a software consistent hashing using an arbitrary key after an NPU node failure; FIG. 7 illustrates a ring virtual configuration of software consistent hashing space of load-balancing groups, construed in accordance with an embodiment of the present invention; and FIG. 8exemplifies a search resolution for a software consistent hashing after packet classification utilizing different load­balancing groups as VRF (Virtual Routing and Forwarding) tables while carrying out a LPM algorithm search.

Claims (7)

1.CLAIMS 1. A method for use in a communication network, wherein said method comprises the steps of:(i) providing a plurality of network processing units (NPU's) comprised in said communication network;(ii) establishing a replication of at least one of said plurality of network processing units;(iii) virtually arranging the plurality of network processing units and the at least one replication, in a ring configuration; (iv) associating a unique primary virtual identification and a corresponding unique backup virtual identification with each active and available entity selected from among said plurality of network processing units and the at least one replication;(v) establishing a list of hash values, each associated with the primary virtual identification or with the backup virtual identification of a corresponding active and available entity;(vi) implementing a ring consistent hashing algorithm for carrying out a search resolution for a consistent hashing; and (vii) in a case of a change in at least one of the active and available entities having a certain unique primary virtual identification, using the corresponding backup virtual identification to maintain the ring continuity.
2. The method according to claim 1, wherein the search resolution for a consistent hashing is conducted based on searching for a packet destination ID using a key built from said packet’s meta-data.
3. The method according to claim 4, wherein packet’s meta­data is derived from the packet's networking headers.
4. The method according to claim 2, wherein the search resolution for a consistent hashing is done by retrieving a respective destination ID while using a Ternary Content- Addressable Memory (TCAM) algorithm.
5. The method according to claim 2, wherein further comprising a step of composing a route space of the networking subnets and/or prefixes, and wherein the search function is based on a Longest Prefix Match (LPM) algorithm.
6. The method according to claim 5, wherein the same search function (LPM) is preserved, and the space is built by using possible virtual identifications as prefixes.
7. The method according to claim 5, further comprising a step of utilizing different load-balancing groups as VRF (Virtual Routing and Forwarding) tables, which in turn are used as parameters while carrying out the LPM algorithm searches, thereby enabling generation of a dedicated consistent hashing space for use as a route space for conveying traffic in the communication network. For the Applicants,By:
IL280293A 2021-01-19 2021-01-19 A method for implementing a consistent hashing in a communication network IL280293A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
IL280293A IL280293A (en) 2021-01-19 2021-01-19 A method for implementing a consistent hashing in a communication network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
IL280293A IL280293A (en) 2021-01-19 2021-01-19 A method for implementing a consistent hashing in a communication network

Publications (1)

Publication Number Publication Date
IL280293A true IL280293A (en) 2022-08-01

Family

ID=82839566

Family Applications (1)

Application Number Title Priority Date Filing Date
IL280293A IL280293A (en) 2021-01-19 2021-01-19 A method for implementing a consistent hashing in a communication network

Country Status (1)

Country Link
IL (1) IL280293A (en)

Similar Documents

Publication Publication Date Title
US10931571B2 (en) Service-function chaining using extended service-function chain proxy for service-function offload
US9627063B2 (en) Ternary content addressable memory utilizing common masks and hash lookups
US9385949B2 (en) Routing controlled by subnet managers
EP2924927B1 (en) Techniques for aggregating hardware routing resources in a multi-packet processor networking system
US9729669B2 (en) Method and system for fast and large-scale longest prefix matching
US11496395B2 (en) Method for implementing a consistent hashing in a communication network
JP2015529061A (en) Fast content routing
US8804735B2 (en) Scalable forwarding table with overflow address learning
WO2018075120A1 (en) Forwarding information base architecture for a hybrid ip and content centric network router
EP2856355B1 (en) Service-aware distributed hash table routing
US7624226B1 (en) Network search engine (NSE) and method for performing interval location using prefix matching
Indira et al. A trie based IP lookup approach for high performance router/switch
Sun et al. DHR-CCN, Distributed Hierarchical Routing for Content Centric Network.
IL280293A (en) A method for implementing a consistent hashing in a communication network
CN115396365A (en) Scale-independent fast route convergence method
Pak Fast packet classification for V2X services in 5G networks
US10205658B1 (en) Reducing size of policy databases using bidirectional rules
Kim et al. Scalable name-based inter-domain routing for information-centric networks
RU2022100956A (en) METHOD FOR CONSISTENT HASHING IN A COMMUNICATION NETWORK
Sun et al. Content-based route lookup using CAMs
Villaça et al. HCube: Routing and similarity search in data centers
Liu et al. Identifier and locator mapping service research
Liu et al. A DHTs-based mapping system for identifier and locator separation network
Anand et al. An efficient mask reduction strategy to optimize storage and computational complexity in routing table lookups
Ni et al. Federated Named Data Networking Testbed for Climate Science