Fully Implicit Parallel Simulation of Single Neurons (Hines et al. 2008)

 Download zip file   Auto-launch 
Help downloading and running models
Accession:97985
A 3-d reconstructed neuron model can be simulated in parallel on a dozen or so processors and experience almost linear speedup. Network models can be simulated when there are more processors than cells.
Reference:
1 . Hines ML, Markram H, Schuermann F (2008) Fully Implicit Parallel Simulation of Single Neurons J Comp Neurosci 25:439-448 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type:
Brain Region(s)/Organism:
Cell Type(s):
Channel(s):
Gap Junctions:
Receptor(s):
Gene(s):
Transmitter(s):
Simulation Environment: NEURON;
Model Concept(s): Methods;
Implementer(s):
{load_file("loadbal.hoc")}
{load_file("binfo.hoc")}

if (name_declared("maxfactor") == 0) {execute("maxfactor = .3")}

proc multisplit() {local c, cm  localobj b, ms, vs, bi, cb, nc, nil
	b = new LoadBalance()
	read_mcomplex(b)
    if (split > 1 || pc.nhost > 1) {
	ms = new Vector(100)
	if (pc.id == 0) {
		c = b.cell_complexity()
		cm = c * maxfactor / pc.nhost
		print "c = ", c, "  maxfactor = ", maxfactor, "  cm = ", cm
		b.multisplit(0, cm, ms)
		//bprint(ms)
	}
	pc.broadcast(ms, 0)
	vs = new VectorStream(ms)
	bi = new BalanceInfo()
	bi.msgid = 1e6
	bi.nhost = pc.nhost
	bi.ihost = pc.id
	bi.bilist.append(new CellBalanceInfo(vs))
	bi.mymetis2(pc.nhost)
	if (pc.id == 0) { bi.stat() }
	cb = bi.bilist.object(0)
	nc = new NetCon(&v(.5), nil)
	cb.multisplit(nc, 100, pc, pc.id)
    }
	//printf("%d cpu complexity %g\n", pc.id, b.cpu_complexity())
	cplx = b.cpu_complexity()
}

// make sure mcomplex.dat exists
proc chk_mcomplex() {localobj f
	if (pc.id == 0) {
		f = new File("mcomplex.dat")
		if (f.ropen()) {
			f.close()
		}else{
			mcomplex()
		}
	}
	pc.barrier()
}		

proc bprint() {local i, i1, i2, i3, n1, n2, n3, np
	np = 0
	i = -1
	printf("%d %g", $o1.x[i+=1], $o1.x[i+=1])
	n1 = $o1.x[i+=1] printf(" %d\n", n1)
	for i1 = 0, n1 - 1 {
		n2 = $o1.x[i+=1]
		np += n2
		printf("  %d\n", n2)
		for i2 = 0, n2 - 1 {
			printf("    %g", $o1.x[i+=1])
			n3 = $o1.x[i+=1]
			printf(" %d\n     ", n3)
			for i3=0, n3 - 1 {
				printf(" %d", $o1.x[i+=1])
			}
			printf("\n")
		}
	}
	printf(" %d pieces\n", np)
}

proc mcomplex() {local i, j localobj f, lb, mt, s
	lb = new LoadBalance()
	lb.ExperimentalMechComplex()
	f = new File()
	f.wopen("mcomplex.dat")
	for j=0, 1 {
		s = new String()
		mt = new MechanismType(j)
		for i=0, mt.count-1 {
			mt.select(i)
			mt.selected(s.s)
			f.printf("%g %s\n", lb.m_complex_[j].x[i], s.s)
		}
	}
	f.close
}

proc read_mcomplex() {local i, j, k localobj f, lb, mt, s, s2
	lb = $o1
	f = new File()
	if (!f.ropen("mcomplex.dat")) { return }
	for j=0, 1 {
		k = 0
		s = new String()
		s2 = new String()
		mt = new MechanismType(j)
		for i=0, mt.count-1 {
			lb.m_complex_[j]
			c = f.scanvar()
			f.scanstr(s2.s)
			mt.select(i)
			mt.selected(s.s)
if (pc.id == 0) if (strcmp(s.s, s2.s) != 0) { execerror(s2.s, " not loaded") }
			lb.m_complex_[j].x[k] = c
			k += 1
		}
	}
	f.close

}

Loading data, please wait...