c++ - 'new' runs out of memory with very little memory allocated -


i'm working on large scientific code run non-interactively on cluster. code uses singleton scheme store large matrices in memory, repeated future multiplications. lately code has been crashing, after allocation of type of these matrices, our exception handler returning output this:

## std::bad_alloc caught in terminatehandler ##########

new ran out of memory.
peak rss memory = 905179136 bytes
current rss memory = 905179136 bytes

#######################################################

each process ought have @ least 16 gb of ram available it, difficult see how running out of memory @ 905179136 bytes; furthermore, manipulating code use more or less memory results in program still crashing, smaller (larger) value 'peak rss memory' reported.

i can think of 2 possibilities:

  1. memory fragmentation.
  2. a large 'new' request being made reason, such exceed 16 gb if serviced.

it may relevant cuda application, , matrices being built on cpu in order sent gpu , deleted. gpu reports no errors, however.

unfortunately, unable reproduce bug in tests, , running within valgrind or gdb crash sufficiently slow (i.e. days of wallclock time) queue system kills job. question how distinguish between these possibilities, or others might there be? in particular there way report size of calls 'new'?


Comments

Post a Comment

Popular posts from this blog

asp.net mvc - SSO between MVCForum and Umbraco7 -

i need integrate sso between mvcforum (forum of company) , umbraco7 (manages cms company). however struggling shared authentication work. tried several ways suggested on other threads still not running expected. here did: 1. in mvcforum (forum.mywebsite.com) 1.1. machine key in web.xml: <machinekey validationkey="the same key umbraco7" decryptionkey="the same key umbraco7" validation="sha1" decryption="aes" /> 1.2. authenticate forms in web.xml <forms name="my-sso-auth" protection="all" path="/" timeout="43200" domain="forum.mywebsite.com" loginurl="/members/logon" enablecrossappredirects="true" /> 2. in umbraco7 (news.mywebsite.com) 2.1. machine key in web.xml: <machinekey validationkey="the same key mvcforum" decryptionkey="the same key mvcforum" validation="sha1" decryption="aes" /> 2.2. a
Read more

Python Tkinter keyboard using bind -

i have problem using bind option can click button in keyboard, , call function once key pressed. tried taking other codes have similar purpose , saw i'm doing pretty same, though still have problem must've missed. thankful if me this. here's code: # -*- coding: utf-8 -*- #imports tkinter import * pil import imagetk, image import time import tkmessagebox #===========================================================================================================================================# #tkinter class class app(frame): def __init__(self, master=none): frame.__init__(self, master) self.pack() #===========================================================================================================================================# #game pieces classes class board: def __init__(self, rows = 7, columns = 6, picture_height = none, picture_width = none): self.rows = rows self.columns = columns self.picture
Read more

ubuntu - Selenium Node Not Connecting to Hub, Not Opening Port -

i'm trying set demo selenium grid. have digital ocean ubuntu 14.02 droplet. (i have 2 , tried set using 2 different machines got nowhere.) i'm trying run hub , node both locally. here input both hub, node, , nmap results. https://gist.github.com/mekhami/d5bbecca5ff0de9d2dc9 so, node starting tcp connection reset , port it's supposed operate on closed. i have had no luck in number of different options , configurations work. but... of course, worked randomly once. don't know why, don't know did, know 1 brief moment, node connected , saw on grid console. tried run test , went down again , i've never seen come up. please let me know more information can provide. update: randomly came online again. don't know happened, might related timeout or time_wait or going on. this because vm/server not have enough entropy available. to resolve problem, install haveged (or other similar things). the instructions of how install haveged on ubu
Read more