HSR — High-availability Seamless Redundancy

What is HSR?

HSR — High-availability Seamless Redundancy

HSR (High-avail­abil­i­ty Seam­less Redun­dan­cy) is a redun­dan­cy pro­to­col for Eth­er­net net­works requir­ing short reac­tion times and high avail­abil­i­ty, as for exam­ple pro­tec­tion sys­tems at elec­tri­cal substations.

Unlike com­mon redun­dan­cy pro­to­cols like RSTP, HSR reacts to any net­work com­po­nent fail­ures seam­less­ly (with­out recov­ery time) and is invis­i­ble to the application.

  • HSR is a net­work redun­dan­cy pro­to­col for Eth­er­net net­works stan­dard­ized by IEC 62439–3:2016 (clause 5)
  • Togeth­er with PRP, it has been adopt­ed in the IEC 61850 stan­dard for sub­sta­tion automation
  • It is most com­mon­ly imple­ment­ed in a ring topol­o­gy
  • It is used in crit­i­cal com­mu­ni­ca­tion net­works that request high avail­abil­i­ty and short switchover times
  • It it is par­tic­u­lar­ly fast, mak­ing it suit­able for appli­ca­tions where the recov­ery time of com­mon­ly used pro­to­cols such as the Rapid Span­ning Tree Pro­to­col (RSTP) is too long
  • It is inde­pen­dent of the appli­ca­tion-pro­to­col and can be used by most Indus­tri­al Eth­er­net pro­to­cols in the IEC 61784 suite
jump to Schema jump to How it works jump to PRP vs HSR

Advantages and Disadvantages of HSR


  • It pro­vides seam­less failover (zero time recov­ery) against fail­ure of any net­work component
  • It is inde­pen­dent of & invis­i­ble to the appli­ca­tion-pro­to­col and can be used by most Indus­tri­al Eth­er­net protocols
  • It can be con­nect­ed with­out a sin­gle point of failure
  • The same nodes can be built to be used in both PRP and HSR net­works, since both rely on the same dupli­cate iden­ti­fi­ca­tion mechanism


  • Laten­cy — Dur­ing the trans­mis­sion of a frame in a HSR ring, each link intro­duces addi­tion­al trans­mis­sion and prop­a­ga­tion delays, while each node caus­es fur­ther pro­cess­ing and queu­ing delays. This can be tack­led by min­i­miz­ing the num­ber of hops.
  • Traf­fic — The trans­mis­sion of dupli­cat­ed copies, which are gen­er­at­ed at the source and cir­cu­late inside the net­work in oppo­site direc­tions, cre­ates addi­tion­al traf­fic. Algo­rithms like QR, VRing can be used to address this issue.
  • Band­width - more band­width is con­sumed due to the dupli­ca­tion of the frames


When to Use HSR

HSR is spe­cial­ly suit­ed for cost-effi­cient net­works with a ring-like topol­o­gy, requir­ing high-speed switchover times close to zero.

How it works

The most com­mon­ly used HSR topol­o­gy is a ring net­work based on one phys­i­cal LAN or two inter­con­nect­ed LANs. The source node dupli­cates the frames and sends them to the des­ti­na­tion node using two dif­fer­ent paths. This means one frame can arrive sev­er­al times, but only the first one is accepted/processed. If one of the paths is bro­ken, due to link or node fail­ure, the frames are still able to reach their destination.

HSR Redundancy Networks Explained

The ter­mi­nal­s/end-nodes are called DANH (Dou­bly attached node using HSR). They must car­ry two Eth­er­net ports to con­nect to the HSR ring and may car­ry addi­tion­al ports for main­te­nance pur­pos­es. They send two dupli­cates of each frame, one to each of the direc­tions in the ring and only accept the first copy when receiv­ing a frame, dis­card­ing the sec­ond. Such nodes can be sen­sors, IEDs (Intel­li­gent Elec­tron­ic Devices), merg­ing units and cam­eras amongst others.

Red­Box­es (Redun­dan­cy Box) are enti­ties to con­nect non-DANH nodes or net­work seg­ments to the HSR ring. They must have at least three exter­nal Eth­er­net ports, two to con­nect to the HSR net­work and one for the inte­gra­tion of the non-HSR nodes. Fol­low­ing the prin­ci­ple of a DANH, Red­Box­es cre­ate and trans­fer dupli­cates of each frame they receive from the out­side of the HSR net­work and send the two dupli­cates to the ring. When they receive a frame, they only for­ward the first copy and remove the second.

prp redundancy schema

PRP vs HSR — Which one is better?

All in all, the choice between HSR and PRP should depend on the spe­cif­ic project, as the project will deter­mine which draw­backs and advan­tages weigh more or less. As a rule, PRP net­work redun­dan­cy is more expen­sive, but also more flex­i­ble than HSR. 

PRP redun­dan­cy usu­al­ly dou­bles the cost of a sin­gle non-redun­dant net­work, which makes the imple­men­ta­tion more expen­sive than many oth­er redun­dan­cy pro­to­cols. A High-avail­abil­i­ty Seam­less Redun­dan­cy (HSR) net­work can also pro­vide the same lev­el of redun­dan­cy and will cost less unless you’re plan­ning on keep­ing replace­ment infra­struc­ture anyways. 

How­ev­er, PRP net­works pro­vide greater flex­i­bil­i­ty in terms of net­work topol­o­gy and allow to con­nect (PRP-) “unaware” nodes with­out a Red­Box (Redun­dan­cy Box), which would be required by HSR. Yet, these singly attached nodes can only com­mu­ni­cate with devices of the same LAN, not tak­ing full advan­tage from the redun­dan­cy. 

Oth­er minor dis­ad­van­tages of HSR are poten­tial tech­ni­cal lim­i­ta­tions in rela­tion to traf­fic, laten­cy and band­width. Still, these prob­lems are not very fre­quent and can be coun­ter­act­ed with dif­fer­ent methods.

Any ques­tions about this?
Ask us!

    I have read and accept the Pri­va­cy Pol­i­cy*

    All per­son­al data that you sub­mit us via this form will be processed by Data Con­troller IGRID, S.L. and saved into CONTACTS pro­cess­ing activ­i­tie, with the pur­pose of answer­ing your inquiry, via email or phone, and for a lat­er fol­low up. Pro­cess­ing shall be law­ful because you give us your con­sent for these spe­cif­ic pur­pos­es. Your per­son­al data will not be dis­closed to third par­ties unless forced by law. You have right to access to, rec­ti­fy, or delete your per­son­al data, and oth­er rights as explained in our Pri­va­cy Policy

    Network Communication

    HSR (High-avail­abil­i­ty Seam­less Redun­dan­cy) is a redun­dan­cy pro­to­col for Eth­er­net net­works requir­ing short reac­tion times and high avail­abil­i­ty, as for exam­ple pro­tec­tion sys­tems at elec­tri­cal substations.

    Unlike com­mon redun­dan­cy pro­to­cols like RSTP, HSR reacts to any net­work com­po­nent fail­ures seam­less­ly (with­out recov­ery time) and is invis­i­ble to the application.

    PRP (Par­al­lel Redun­dan­cy Pro­to­col) is a redun­dan­cy pro­to­col for Eth­er­net based net­works requir­ing high avail­abil­i­ty and a short switchover time, as for exam­ple pro­tec­tion sys­tems at elec­tri­cal substations.

    Unlike com­mon redun­dan­cy pro­to­cols like RSTP, PRP reacts to any net­work com­po­nent fail­ures seam­less­ly (with­out recov­ery time) and is invis­i­ble to the application.

    Eth­er­net is a fam­i­ly of wired com­put­er net­work­ing tech­nolo­gies com­mon­ly used in local area net­works (LAN) and also wide area net­works (WAN).

    Over time, Eth­er­net has large­ly replaced com­pet­ing wired LAN tech­nolo­gies by pro­vid­ing high­er bit rates, a greater num­ber of nodes, and longer link dis­tances and decent back­ward compatibility.

    The Medi­um Access Con­trol (MAC) sub­lay­er pro­vides flow con­trol and mul­ti­plex­ing for the trans­mis­sion medi­um to con­trol the hard­ware that inter­acts with the wired, optic and also wire­less trans­mis­sion media in the IEEE 802 LAN/MAN data link layer.

    The MAC is accom­pa­nied by the LLC sub­lay­er, which pro­vides flow con­trol and mul­ti­plex­ing for the log­i­cal link (i.e. Ether­Type, 802.1Q VLAN tag etc.)

    Vir­tu­al Local Area Net­work (VLAN) is a is a sub­net­work which can group togeth­er col­lec­tions of devices that are con­nect­ed to sep­a­rate phys­i­cal LANs.

    VLANs allow net­work admin­is­tra­tors to par­ti­tion a sin­gle switched net­work in order to keep net­work appli­ca­tions sep­a­rate despite being con­nect­ed to the same phys­i­cal net­work, with­out requir­ing new cabling or major changes in the cur­rent net­work infrastructure.

    Your con­tent goes here. Edit or remove this text inline or in the mod­ule Con­tent set­tings. You can also style every aspect of this con­tent in the mod­ule Design set­tings and even apply cus­tom CSS to this text in the mod­ule Advanced settings.

    Net­work redun­dan­cy is a method to ensure net­work avail­abil­i­ty, pro­vid­ing failover when a device or net­work path fails or becomes unavailable.

    Redun­dan­cy is usu­al­ly achieved by installing addi­tion­al or alter­na­tive net­work devices, com­mu­ni­ca­tion media or equip­ment with­in the net­work infrastructure

    all entries sort­ed aplhabetically

    A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

    A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

    HSR & iGrid

    Redun­dan­cy has been a key ele­ment in many iGrid projects, which is why all of our devices sup­port var­i­ous redun­dant schemes and net­works, includ­ing PRP, HSR and RSTP.

    iRTU/iGW‑S#01 devices can acts as HSR DANH and Red­Box (Redun­dan­cy Box).

    iRTU – With I/Os for Direct Data Acquisition 

    Com­pact and scal­able bay con­troller which can act as IEC 61850 client or serv­er, fea­tur­ing con­fig­urable I/O boards for direct data acqui­si­tion, high-pre­ci­sion time­stamp­ing and an option­al Eth­er­net switch for addi­tion­al Eth­er­net ports.

    iGW‑S Substation Gateway

    Pow­er­ful and reli­able sub­sta­tion gate­way, able to run either in stand­alone or redun­dant modes, with an embed­ded Eth­er­net switch (4 ports) and IEC 61850 client and serv­er capabilities.

    iGrid Solutions and Applications

    Automation with IEC 61850 

    The IEC 61850 stan­dard is enabling new opor­tu­ni­ties for ven­dor inter­op­er­abil­i­ty and advanced sub­sta­tion automa­tion. Find out how you can take advan­tage of IEC 61850 with easy-to-use and adapt­able solu­tions for a sim­ple migra­tion or retrofit.

    HV Substation Automation

    Pow­er­ful sub­sta­tion automa­tion sys­tems often han­dle numer­ous com­mu­ni­ca­tion pro­to­cols and media with­in one net­work, which can result in expen­sive and com­plex projects.  Avoid these prob­lems with inter­op­er­a­ble tech­nol­o­gy and smart con­fig­u­ra­tion tools.

    MV Distribution Grid Automation

    It is often dif­fi­cult to find the exact solu­tion you need in a MV appli­ca­tion, lead­ing to high­er costs than nec­es­sary. With our scal­able and adapt­able solu­tions you will be able to only pay for what you real­ly need, with­out com­prim­is­ing on qual­i­ty or security.

    Photovoltaic Power Station

    Using an open and scal­able SCADA sys­tem to mon­i­tor and con­trol a PV plant comes with many ben­e­fits on sev­er­al lev­els. Find out how advanced com­mu­ni­ca­tion tech­nol­o­gy affects PV oper­a­tion, main­te­nance, sys­tem design, invest­ment secu­ri­ty, profits…

    Protocol Conversion

    As com­mu­ni­ca­tion net­works grow in com­plex­i­ty, “plug and play” promis­es become hard­er to keep. Inter­op­er­a­ble pro­to­col con­vert­ers and soft­ware solu­tions with state-of-the-art capa­bil­i­ties and funci­tonal­i­ties can be the bridge to all the func­tions and flex­i­bil­i­ty your net­work needs.

    Generation Dispatch Control Center

    With a gen­er­a­tion dis­patch enter you can auto­mat­i­cal­ly con­trol the gen­er­a­tion of all pow­er plants and make direct bids for ancil­lary ser­vices on one plat­form. Check out the most effi­cient com­mu­ni­ca­tion path between gen­er­a­tion sites, grid oper­a­tors and the pow­er market.

    Smart Metering

    A sin­gle device that col­lects, process­es, trans­fers smart meter data and load curves from sev­er­al meters in dif­fer­ent pro­to­cols via ser­i­al or Eth­er­net, whilst pro­vid­ing advanced automa­tion func­tions? Adapt­able designs and a full com­mu­ni­ca­tion pro­to­col suite make it possible. 


    Switchgear & Transformers

    Some­times you have pre­ferred gear for a project or it has already been installed, but it is lack­ing the com­mu­ni­ca­tion capa­bil­i­ties to pro­vide the automa­tion func­tions you are look­ing for. With our soft­ware core iGComms any device can be as smart as you want it to be.