Connect:    

FAQ

General   |   MILC   |   WRF   |   GROMACS   |   POWER

Last updated: June 09, 2013

GENERAL


The FAQ section contains the frequently asked questions and answers that are applicable for all student team. Please watch this page for updates. To submit questions about the Student Cluster Challenge, please fill out the web form below.

Q1. Will we be allowed to use hardware or software that has not yet been publicly or commercially released?

Answer:
You are free to use any kind of hardware or software, as long as the power for the hardware is from the given budget – you are not allowed to generate power in any way or form on top of the given power budget.

Q2. Is modification of the source code , operating system or hardware allowed?

Answer:
You are allow to do any modifications, again, as long as you are not using more than the given power budget.

Q3. What is all included in the power limit? Would a desktop pc controlling the cluster or a nagios monitoring box be included?

Answer:
A laptop/desktop machine that does system monitoring is ok not to include in the power budget. The power budget covers all of the equipment that is actively used for the applications and benchmarking (for example servers. Monitors for example is not part of the budget.

However, if you plan on using a system provisioning software with management, that is responsible for distributing the software images/packages to the compute nodes. That will count as a part of the cluster and should count as a part of the power budget.

The rule is, if it has any direct impact on the application performance, and we need to account that for the power budget.

One exception: Using external/stored power source is not allowed. Such as using battery backup (for instance, Uninterruptible Power Supply or UPS) is not allowed.

Examples of power usage that counts toward power budget:
- servers
- network switches
- system provision nodes
- laptop/desktop running system monitoring software for provisioning system images.
- File system server (such as NFS server)
- Fans, and other method of cooling (such as liquid cooling)

Examples of power usage that does not count towards power budget:
- laptop/desktop running system monitoring software for the power usage.
- monitor/screen/KVM

Q4. The two secret benchmarks are presented on the first and second day, when do we have to hand in the results for these benchmarks?

Answer:
There will be one secret benchmark on the first day, and another given on the second day. You will be provided with the needed information per application at the morning of each of the days.

Q5. Do we need any special version of any program to run the benchmarks. Can we use the DEISA benchmark suite?

Answer:
You may try to run and discover any type of benchmark dataset and suites that are available for your practice purposes. As for preparation, We expect teams to build the applications and run a few benchmarks to get familiar with the application and dataset to discover the characteristics of an application and datasets to their own advantages.

Q6. What is the order of magnitude of the benchmarks' runtime? Minutes or hours?

Answer:
It will be a combination

Q7. In which form are the benchmark inputs provided?

Answer:
Tentatively speaking, the input files will be given in a USB drive to each team at the beginning of each of the days, along with the mystery application of the day and any special instructions on running the applications dataset.

Q8. How are the results compared? What kind of tolerance is there. When is a result considered unacceptable or correct? Is it for instance allowed to calculate some parts in single precision?

Answer:
Specific instructions on what floating point precision the application needs to run will be given along at the time when applications are handed out at the beginning of the day.

Q9. Is it correct to suppose that the students do not have to pay the registration for the ISC? What about the supervisor?

Answer:
Students who participate at the Student Cluster Challenge will not need to pay for the registration for ISC. Supervisor will not have to pay.

Q10. Is it allowed to change the BIOS settings (either change in the BIOS menu, in EFI shell or in OS-level tools) and reboot in the systems during the competition?

Answer:
No changes in the BIOS are allowed after the competition has started. The systems are to be powered on at all times and no rebooting. System reboots can only reboot when the system is hung or it has encountered major failures.

Q11. I was on the site not specified the size of the input files are for the applications. Would be good to know, however, so you can choose the size of the disks accordingly? Can you make as an estimate?

Answer:
The input files will be handed to each team using a storage media, for example USB(s) or DVD ROM(s), before the competition starts on each day. For some applications, even the input files may be small, there are steps involved, such as domain decomposition, which will create additional data files for each CPU to process. Thus generate much more space than just the input files that are given. Teams will need to be familiar with the benchmark applications to understand the typical data set and the runtime IO requirements in order to come up with a good enough disk storage solution.

Q12. Regarding the "Powered On at All Times and No Rebooting" rule, do the following actions break the rule?

Answer:
Modes like hibernation or suspension (either suspend-to-RAM or to disk) are not allowed. Those modes do not qualify for the rule "power on at all times and no rebooting" because power on mode implies that compute nodes need to be in the operating mode throughout the competition. Hibernation, suspension, and sleep modes are not in the operating mode, they are in the Standby mode, thus not qualify according to the rule.

Q13. Which version of the given applications we are supposed to run at the competition? Because applications contain set of applications and utilities in them which differ from version to version.

Answer:
You can use a recent version or the latest version of the code at the competition.

Q14. Could you please create a link to our blog on our teams page on the website of the SCC?

Answer:
Absolutely. If you do have additional info that you would like to add to the ISC'13 SCC page, please let us know. We have plans to promote the web pages by including more personalized profiles as well. This helps to drive more interests and awareness to the competition. We are updating the ISC'13 SCC pages continuously.

Q15. Do I need to register the students participating in the competition for the ISC conference or with that be done through the HPCAC?

Answer:
All competition members including group leaders will receive a full pass for the conference. The registration codes will be provided in April.

Q16. What are the expectations for mounting hardware at the competition? Our nodes are contained in a single chassis, plus the admin server. Is it required for us to put this gear in a rack, or can it be left on a table since the nodes aren't loose.

Answer:
No it is not required. The systems can be put in a rack, on a table, on the floor, or anywhere in the booth. Some teams attempted with non-traditional (e.g. liquid) cooling so the systems might not be racked. It is for the team to decide.

Q17. Regarding logistics in Germany, is there bus or train service between hotels and conference center?

Answer:
There are information about traveling to the convention center (CCL) on this web page. http://www.isc-events.com/isc13/travel.html According to the ISC’13 Travel page, this year the ISC registration fee includes travel within Leipzig (zone 110), so, please wear your badge when alighting any public transportation from June 16 until June 20. For detailed tram schedules and information can be found at: http://www.lvb.de

Q18. Do I need to register the students participating in the competition for the ISC conference or with that be done through the HPCAC?

Answer:
All competition members including group leaders will receive a full pass for the conference. The registration codes will be provided in April.

Q19. Can you describe how the overall points were giving for the competition this year?

Answer:
The Linpack performance is 10% of the score, applications will add 50% and the last 40% from the interview.

Q20. What will the internet connectivity be like at the venue? Will cabled internet access be provided for the teams?

Answer:
There will be access to internet at the venue. There will be 2 wired Ethernet cables for each team's booth. One is to connect to the AP8858 PDU. Another cable is for connected to the team's wired router for their own internal use. No own wireless router is allowed at the venue. And finally there will be wifi available but the connectivity of that may be unreliable depending on the utilization by the attendees at the venue.

Q21 . As a mentor I am not allowed to help the team during the competition. Can I help them during the setup period before the competition starts?

Answer:
Mentor can help during setup period.

Q22 . Does a server running DNS, DHCP and a probably a batch system counting to the power budget?

Answer:
DNS and DHCP can be on a separate power that does not count toward power. It eases the management of the systems and does not affect application performance.

Q23 . Do we have some kind of internet connection for the cluster and for the laptops of the students? Is it a global network infrastructure with the risk of attacks? Can we bring some (soho) router with Wifi connection? Maybe it is necessary to install some packages and so on.

Answer:
There will be 2 Ethernet cables provided to each team’s booth. One cable is to connect to the PDU for monitoring the power draw by the organizer. Another has the access to the internet. The teams can connect the internet cable to their internal router for Internet connectivity. However, only wired router is allowed; the teams cannot bring wireless router that serves wifi connectivity due to the restriction by the venue (CCL).

MILC


Q1. Which exact applications we are supposed to run at the competition? e.g. Previous version of MILC 7.7.5 does not have application 'arb_overlap'. Recent version 7.7.8 contains about 18 applications in it.

Answer:
The “ks_imp_rhmc” application will be used. You would need to use the latest version of MILC.

Q2. Some applications like 'arb_overlap' in the MILC release 7.7.8 have broken code in the release. Are we supposed to make changes to the code?

Answer:
We are not aware of the broken code in the application. If you found issues with the code, we would encourage you to contact the authors or the developers for the open source application, or utilize their mailing alias to ask questions. You can also send specific questions and we can forward the questions to the developers. You may make changes to the code.

Q3. In the application code for the MILC 7.7.8 we found 3 applications as well which are creating executable file su3_rmd. Could you please let us know which of them or all of them we need run as part of benchmarks?

Answer:
The “ks_imp_rhmc” application will be used. There is a target called “su3_rhmc_hisq” which is for using the HISQ algorithm.

WRF


Q1. Which version of WRF should be used?

Answer:
The teams should use WRFV3.5 from NCAR but it needs a couple of changes/bug-fixes. These have been incorporated into the NCAR repository but not released to the users yet (they'll come out in WRFV3.5.1).

  1. In the configure.wrf file that gets generated for the x86_64 i486 i586 i686, Xeon (SNB with AVX mods) ifort compiler with icc  and MIC option, there is a setting, -DINTEL_ALIGN64,  that needs to be removed from the ARCH_LOCAL setting.   This will cause performance degradation if it's there.
  2. There is a bug in code that's exercised by this case.  It can cause occasional (and non-reproducible) segmentation violations. The fix is to the file frame/module_domain.F in the WRFV3.5 distribution. Click here to see the svn diff output.

GROMACS


Q1. Will the Gromacs benchmark require single or double precision? This is interesting for us because the GPU acceleration of Gromacs is not compatible with double precision.

Answer:
Gromacs benchmark will most likely be single precision. But I cannot confirm it at this time because the benchmark dataset used at the SCC is not yet finalized.

POWER


Q1. How is the power provided? One PDU with all of the power limit or two PDUs with half of the power limit each?

Answer:
The new power distribution unit (PDU) supports drawing a maximum of 16A. We are using the APC AP8858 PDU for each team to monitor the power usage through its Ethernet interface with SNMP. The TypeF wall plug from the conference floor can supports up to a maximum of 16A power. The team will use the power cap at 3000W (~to 3kW). At the competition booth, there will be an adapter from the typeF wall power to the C20 socket for the input of the PDU.

http://www.apc.com/resource/include/techspec_index.cfm?base_sku=AP8858

There are 18 outlets for C13 and 2 outlets for C19 available. If more devices are needed to be plugged into the PDU to account for the power budget, the team will need to prepare and bring enough splitter cables (IEC C14 to C13 Splitter power cables) to power the additional devices. If the type of the plugs used for the machines are not taking C13, the teams will be responsible for bringing suitable power cables.

Q2. What exactly happens if you take exceeds the power limit? If we disqualified now? Or do you get negative points? Or you can simply try again?

Answer:
We will collect power consumption from the PDU to monitor the power usage by the teams. As soon as the power consumption is over 3000W, the overload warning light on the PDU will be illuminated to notify the team has exceeded the power budget. The SCC supervisors will be notified too that the power exceeded the limit. Points will be deducted each time the power budget is exceeded.

If the power consumption get well beyond the 3000W limit, Bad Things™ will happen and the circuit breaker will trip and you will lose power to the systems. The team will also lose valuable system time to run the dataset while the circuit breaker is being reset, so the team will need to control the power draw throughout the competition.

Q3. What is provided in the competition place? Is there a LED monitor or we shall take our own?The AP8858 PDU will provide output for up to 8 devices. If more devices are needed to be plugged into the PDU to account for the power budget, can we prepare a power strip which will be plugged into the PDU?

Answer:
There will be 2 Type F power outlets (at least one of the outlets is rated at 16A) will be provided for each team. Only one PDU will be given to each team which will be plug into 1 of the power outlets. The other Type F power outlet will have a power strip. Each team will be responsible for bringing in accessories, such as power strip (that takes the Type F plug), power cables for electric equipment, computer monitors, splitter cables (with proper plugs), etc. If you need to connect more devices, the team will need to bring in additional power strip. Please note that computer monitor can be excluded from the power budget.

Q4. About the 3kW power limit, we want to know the detail rules, i.e. when the cluster's power is over 3000W, would the power be turn down immediately or there is a little time before the power turned down? Points will be deducted each time the power budget is exceeded how many each time?

Answer:
The AP8858 PDU will be set to trigger a warning light on the PDU. As soon as the cluster power exceeds 3000W during the actual competition, the team will see the overload warning light on the PDU will be illuminated to notify the team has exceeded the power budget. The SCC supervisors will be notified too that the power exceeded the limit. Points will be deducted each time the power budget is exceeded.

SCC supervisors may disqualify one or more run submissions if they observe that the power consumption is beyond the limit for an extended period of time. The idea is for teams NOT to exceed the 3000W limit. The team needs to control the power draw to 3000W or less.

If the power consumption reaches around 3000W, Bad Things™ will happen and the circuit breaker will trip and you will lose power to the systems. The team will also lose valuable system time to run the dataset while the circuit breaker is being reset, so the team will need to control the power draw throughout the competition.

Q5. The PDU APC AP8858 (used last year) measures only current but not power. So a fluctuating voltage changes the available power but is not recognized by the PDU. Do you plan on using another PDU for SCC 13?

Answer:
The new APC AP8858 PDU which will be used this year will fix one of the limitations by the old PDU which can only report back the current (and not to account for the voltage fluctuation due that affects power when power draw is under heavy load). The new AP8858 PDU also provides more output receptacles which allows more devices to be plugged in.

 


Submit your question here:

* Name:
* Email:
* University:
* Question:
* What is 3 + 6?