we’re looking forward to training some information on how remanufacturing applies to IT and how this differs from mechanical remanufacturing so we found a lot of the public exception is the IT hardware degrades over time and so you see a performance decrease of remanufactured against original products the xfurbish of East London recently carried out research that proves remanufactured service perform identically to new in fact that the immediate previous generation can outperform.

The latest if it’s configured correctly so joining me on this this presentation this webinars is teal she’s the research associate from the University of East London hi everyone so before we start with the research I’m going to explain a little bit about our manufacturing process which is covered on this slide so tech buyers a global expert in buying refurbishment and selling of enterprise IT.

Servers storage and networking for the data enter environment or we’d classes Enterprise spares a lot of our business is focused on bringing in redundant IT from a variety of sources we execute a 25-point testing program on servers data sanitization where necessary and everything is restored to an as new condition to be sold as spares and upgrades we also have a remanufacturing portion of the online business which rebuilds fully configured service.

Which is that with a client according to individual client needs from the  component parts and installs the  relevant software everything gets issued  a three year warranty on all our  products anything we can’t use ghost  recycling but everyone in the business  has a genuine hatred of waste that’s company-wide if we can reuse or  repair we do that as a preference every  single time our technical staff repair components wherever possible and our  expertise and this is growing so from sort of memory and tape drives and all  sorts of things I would say our customers care less about that though  than the economic benefits of remanufacture so we had their paper  written by a post-grad one of our local business schools that said the customers  buy from us because of cost and  performance rather than the  environmental considerations the  standard argument against using  remanufactured equipment in the data centre is that new equipment is more  energy efficient since the IT load  for the biggest proportion of energy use  in the sector it’s important for  customers consider that in the past new  service significantly increase the  amount of compute power per watt so the  advice was janitor refresh often and to refresh with new because there was  significant benefits on the latest generation however as times.

New generations are making less of an  improvement on older models this is for  a variety of reasons and this is what  our study with the University of East London was designed to explore so I’ll  hand over to Noor now to take you  through it thanks rich so a server is in  many ways a large computer if you’re  open wound up you will see a motherboard  CPUs which are behind the heat sinks in  this picture memory storage transistors  and fans to keep everything from  overheating if we stirred this down two  essential components we have the CPU  which gives the processing power memory  which provides short-term access to data  and storage which you can think of as  the server’s filing cabinet these are  the components we concentrated on during  our experiments when most people think  of the compute power of a server they  think the CPU is the workhorse it is  generally front and centre of all  discussion around compute performance  however there are two helpers in this  function storage and memory that can be  very impactful if you add more storage  or more memory you can upgrade the  machine we are about to show you what  that looks like a big part of the study  was understanding server trends over  time a study was conducted by uptime  Institute which analyses performance data.

 for hundreds of servers between  2007 and 2013 the graph displays how CPU  technology evolved over time along with  performance and idle power consumption  the figure reveals interesting insights  during the beginning of the decade the  move from one CPU lithography to another  for example 65 nanometres to 49  nanometres presented major performance  per Watts gains represented in the  orange line as well as substantial  reduction idle power shown in the blue  law however over the past few years  while lithography was fixed at 40  nanometres the increase in performance  per watt has been escorted by a study  increased an idle power consumption  perhaps due to the increase in core  count to achieve performance gains this  is one reason why the case for hardware  refresh for more recent get has become  weaker especially that servers and  real-life deployments tend to spend 75  percent of their time on average an idle  mode as such the increase in idle power  may offset energy gains from performance  it was of great interest for us to study  the performance gain between latest  server generation and an older  generation released after 2015 we  expected the gain to be minimal due to  the server trend discussed above  therefore remanufacturing machines from  a previous generation could be a good  answer for energy efficiency as well as  saving on the vital precious metals and  critical raw materials that IT hardware.

Contains the first stage was to verify that a new and refurbished server  against the same make and model had the same level of performance this is done  by measuring energy draw and performance  measured as number of operations done  per second of the CPU storage memory as  well as the overall score  we then tested different combinations of  new and refurbished for example new machines with refurbished components and  refurbished machines with new components our results showed that there was no  real difference between the two first this was not a surprise given the last  moment service is much longer than 3 years.

However it was good to have this verified by university to the best of  our knowledge this is the first time  this type of research had actually ever  been done we really wanted to know what  would happen when an older server was  upgraded with additional capacity we  would see if it could outperform the  last generation that was covered by the  next stage of testing the showcase  comparison is between an HPE ProLiant  dl380 gen9 which was released in quarter  4 in 2014 and an HPE ProLiant dl380 gen9  released in quarter for 2017 these were  chosen since they are very popular  servers but we plan to expand on this  testing for more makes and models we run  the same set of tests as we did for  comparing new and refurbished energy  efficiently but this time varied the  components and the older generation 9  model we found that a generation 9 with  a 53 and 54 duo CPUs outperformed a generation 10 model with base configuration using dual sky lake silver cpus just by adding more ram the results  verified that the performance gain  between latest generation and previous  one was in fact very little as discussed earlier on we’d like to add though.

This would had not been the case for  servers released prior to mid-2014  because server performance jumped  significantly after which but we do  expect this to be the case for the next  few years until a breakthrough and CPU  technology takes place now these tests  were part of the longer-term projects  where we building a data center energy  efficiency modeling tool to provide you  with scientific recommendations to  improve your overall data center  efficiency primarily by improving server  hardware configuration however what is  most interesting here is that we’ve been  able to get the immediate previous  generation two out for latest model  that’s what’s demonstrated on this slide  at the moment for a long time the  industry believed you Apso.

We had to refresh with the new and  latest in order to achieve maximum  efficiency what this proves is that in  certain environments remanufacture  machines could be the better choice to  cost performance and energy drop as I mentioned at the beginning the  presentation cost and performance are  obviously really important many  organizations most people they care  about the bigger picture of cutting  waste and pollution as well as  protecting our natural resources the exact size of the environmental impact  is under debate at the moment different  experts will tell you different things  we’ve tried to keep this this slide in  line with this we do know that both  waste and energy usage is on a steep  upward curve for the data center industry so resource and waste and  reduce air emissions are all increasing  so this is the latest information.

That we have on what it costs us to make a  new server it’s from a Deloitte report in 2015 which informed the eco design directive so it’s kind of the best  information available at the time tech buyer is currently involved with a  multinational research project called sedate which will update and improve on  this information the likelihood is that  modern servers have a higher carbon  footprint and water usage the nice  artistic suggests but as you can see on  the slide here this there’s quite a lot  of knowable materials and energy gone  into the actual production of servers  the macro picture on this is that  humanity is spending way over its limit  in terms of its use of natural resources  and also what the earth can recover from  by 2050 they say we spend will be three  times the amount the planet can provide  that’s unless we see we change how we do  things this kind of slide talks about  the ways we can make best use of  resources but the order is important too  so if we maintain our assets for as long  as possible we use a hundred per set of  materials and use minimum additional  energy refurbishment which is restoring  goods to factory condition and reuse  which is building them into new machines  is our next best option in our sector  this is what remanufacturing means there  are limits to which components go with  which so remanufacturing servers.

Sticking within the confines of each-generation but we can increase performance within these parameters the last and final option is to recycle especially with ICT this is because many of the materials exist in trace amounts and not all of them are recoverable so you’ll never get 100% return rate also with recovering the materials you have an additional energy cost.

Which you  want to avoid if you can this kind of  thinking comes in the heading circular  economy now although the remanufacturing  industry existed long before the term  was invented those that support circle  economy in the data centre industry  point to the Google model to show that  it can be economically as well as  environmentally successful so you know  looking at the slide some of the key  points on this one is that they’ve kind  of been refurbishing their instance 2015  which is a long time they don’t  just make a distinction between  refurbished and new inventory both are  considered equivalent so they both sit  on the shelf and they’re not separately  labelled and you know it saved Google  about billion dollars in operations to  energy efficiency alone by kind of  refurbishing its own computing  infrastructure now Google is able to do  this because it manufactured its own servers and control the outgoing and  return supply chain so what companies we do is provide smaller organizations with the capacity to do  the exact same thing by effectively  working as one of their service arms  it’s a little different financially  because we buy equipment in and then  resell but this works well for our  customers because they don’t have the  overhead they’re also probably not big  enough to do on a rolling basis so they  get the benefit of specialist  remanufacture perfected over many years  even if they only use it once so we know  that the remanufactured service perform  as well as new and that it’s possible to  get the immediate past generation to  outgun the latest generation with the  correct component level upgrades so what  does that mean for remanufacturing as a  whole we believe it means taking a  different approach to IT remanufacturing  rather than mechanical remanufacturing  for IT we can take customers to a  decision tree where questions these are asked on the slide you know  there is no wear and tear for IT in the  same way that there is for mechanical  components you know in fact previous  models can outperform newer models with  the right component upgrades.

What you’re buying, what it’s gonna be used for and where it’s going is this going to be  production is it going to be test is it  going to be dev we know is it a noose  and you know look at ways you can  prolong the life of existing equipment  by using at great upgrades of components  rather than complete refreshing and also  look at the cost saved  and what you could buy by saving that so  this could be security software a  critical infrastructure it could be  off-site backup solutions there’s a lot  stuff you could be spending the money on  better than spending it on new kit that  you don’t need so I guess you know  what’s the point of all this  well mostly that it makes sense and if  you have a product that cost less  performs just as well and is good.

By Anurag Rathod

Anurag Rathod is an Editor of, who is passionate for app-based startup solutions and on-demand business ideas. He believes in spreading tech trends. He is an avid reader and loves thinking out of the box to promote new technologies.