-
\documentclass[smallextended]{svjour3}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{mathptmx}
+\usepackage{amsmath}
\usepackage{courier}
\usepackage{graphicx}
+\usepackage[ruled,lined]{algorithm2e}
+
+\newcommand{\abs}[1]{\lvert#1\rvert} % \abs{x} -> |x|
+
+\newenvironment{algodata}{%
+ \begin{tabular}[t]{@{}l@{:~}l@{}}}{%
+ \end{tabular}}
+
+\newcommand{\VAR}[1]{\textit{#1}}
\begin{document}
-\title{Best effort strategy and virtual load for asynchronous iterative load balancing}
+\title{Best effort strategy and virtual load
+ for asynchronous iterative load balancing}
\author{Raphaël Couturier \and
Arnaud Giersch \and
Abderrahmane Sider
}
-\institute{F. Author \at
- first address \\
- Tel.: +123-45-678910\\
- Fax: +123-45-678910\\
- \email{fauthor@example.com} % \\
-% \emph{Present address:} of F. Author % if needed
+\institute{R. Couturier \and A. Giersch \at
+ LIFC, University of Franche-Comté, Belfort, France \\
+ % Tel.: +123-45-678910\\
+ % Fax: +123-45-678910\\
+ \email{%
+ raphael.couturier@univ-fcomte.fr,
+ arnaud.giersch@univ-fcomte.fr}
\and
- S. Author \at
- second address
+ A. Sider \at
+ University of Béjaïa, Béjaïa, Algeria \\
+ \email{ar.sider@univ-bejaia.dz}
}
\maketitle
is certainly the most well known algorithm for which the convergence proof is
given. From a practical point of view, when a node wants to balance a part of
its load to some of its neighbors, the strategy is not described. In this
-paper, we propose a strategy called \texttt{best effort} which tries to balance
+paper, we propose a strategy called \emph{best effort} which tries to balance
the load of a node to all its less loaded neighbors while ensuring that all the
nodes concerned by the load balancing phase have the same amount of load.
Moreover, asynchronous iterative algorithms in which an asynchronous load
balancing algorithm is implemented most of the time can dissociate messages
concerning load transfers and message concerning load information. In order to
increase the converge of a load balancing algorithm, we propose a simple
-heuristic called \texttt{virtual load} which allows a node that receives an load
+heuristic called \emph{virtual load} which allows a node that receives a load
information message to integrate the load that it will receive later in its
load (virtually) and consequently sends a (real) part of its load to some of its
neighbors. In order to validate our approaches, we have defined a simulator
no external load. In this context, Bertsekas and Tsitsiklis have proposed an
algorithm which is definitively a reference for many works. In their work, they
proved that under classical hypotheses of asynchronous iterative algorithms and
-a special constraint avoiding \texttt{ping-pong} effect, an asynchronous
+a special constraint avoiding \emph{ping-pong} effect, an asynchronous
iterative algorithm converge to the uniform load distribution. This work has
-been extended by many authors. For example,
-DASUD~\cite{cortes+ripoll+cedo+al.2002.asynchronous} propose a version working
-with integer load. {\bf Rajouter des choses ici}.
+been extended by many authors. For example, Cortés et al., with
+DASUD~\cite{cortes+ripoll+cedo+al.2002.asynchronous}, propose a
+version working with integer load. This work was later generalized by
+the same authors in \cite{cedo+cortes+ripoll+al.2007.convergence}.
+{\bf Rajouter des choses ici}.
Although the Bertsekas and Tsitsiklis' algorithm describes the condition to
ensure the convergence, there is no indication or strategy to really implement
the load distribution. In other word, a node can send a part of its load to one
or many of its neighbors while all the convergence conditions are
-followed. Consequently, we propose a new strategy called \texttt{best effort}
+followed. Consequently, we propose a new strategy called \emph{best effort}
that tries to balance the load of a node to all its less loaded neighbors while
ensuring that all the nodes concerned by the load balancing phase have the same
amount of load. Moreover, when real asynchronous applications are considered,
migrates from one node to another one. Depending on the application, it may have
sense or not that nodes try to balance a part of their load at each computing
iteration. But the time to transfer a load message from a node to another one is
-often much nore longer that to time to transfer a load information message. So,
+often much more longer that to time to transfer a load information message. So,
when a node receives the information that later it will receive a data message,
it can take this information into account and it can consider that its new load
is larger. Consequently, it can send a part of it real load to some of its
-neighbors if required. We call this trick the \texttt{virtual load} mecanism.
+neighbors if required. We call this trick the \emph{virtual load} mechanism.
So, in this work, we propose a new strategy for improving the distribution of
the load and a simple but efficient trick that also improves the load
-balacing. Moreover, we have conducted many simulations with simgrid in order to
+balancing. Moreover, we have conducted many simulations with SimGrid in order to
validate our improvements are really efficient. Our simulations consider that in
order to send a message, a latency delays the sending and according to the
network performance and the message size, the time of the reception of the
and Tsitsiklis' asynchronous load balancing algorithm. Moreover, we present a
possible problem in the convergence conditions. Section~\ref{Best-effort}
presents the best effort strategy which provides an efficient way to reduce the
-execution times. In Section~\ref{Virtual load}, the virtual load mecanism is
+execution times. In Section~\ref{Virtual load}, the virtual load mechanism is
proposed. Simulations allowed to show that both our approaches are valid using a
quite realistic model detailed in Section~\ref{Simulations}. Finally we give a
conclusion and some perspectives to this work.
\label{BT algo}
In order prove the convergence of asynchronous iterative load balancing
-Bertesekas and Tsitsiklis proposed a model
+Bertsekas and Tsitsiklis proposed a model
in~\cite{bertsekas+tsitsiklis.1997.parallel}. Here we recall some notations.
Consider that $N={1,...,n}$ processors are connected through a network.
Communication links are represented by a connected undirected graph $G=(N,V)$
-where $V$ is the set of links connecting differents processors. In this work, we
+where $V$ is the set of links connecting different processors. In this work, we
consider that processors are homogeneous for sake of simplicity. It is quite
easy to tackle the heterogeneous case~\cite{ElsMonPre02}. Load of processor $i$
at time $t$ is represented by $x_i(t)\geq 0$. Let $V(i)$ be the set of
When a processor send a part of its load to one or some of its neighbors, the
transfer takes time to be completed. Let $s_{ij}(t)$ be the amount of load that
-processor $i$ has transfered to processor $j$ at time $t$ and let $r_{ij}(t)$ be the
+processor $i$ has transferred to processor $j$ at time $t$ and let $r_{ij}(t)$ be the
amount of load received by processor $j$ from processor $i$ at time $t$. Then
the amount of load of processor $i$ at time $t+1$ is given by:
\begin{equation}
Some conditions are required to ensure the convergence. One of them can be
-called the \texttt{ping-pong} condition which specifies that:
+called the \emph{ping-pong} condition which specifies that:
\begin{equation}
x_i(t)-\sum _{k\in V(i)} s_{ik}(t) \geq x_j^i(t)+s_{ij}(t)
\end{equation}
for any processor $i$ and any $j \in V(i)$ such that $x_i(t)>x_j^i(t)$. This
-condition aims at avoiding a processor to send a part of its load and beeing
+condition aims at avoiding a processor to send a part of its load and being
less loaded after that.
Nevertheless, we think that this condition may lead to deadlocks in some
cases. For example, if we consider only three processors and that processor $1$
is linked to processor $2$ which is also linked to processor $3$ (i.e. a simple
-chain wich 3 processors). Now consider we have the following values at time $t$:
+chain which 3 processors). Now consider we have the following values at time $t$:
\begin{eqnarray*}
x_1(t)=10 \\
x_2(t)=100 \\
In this case, processor $2$ can either sends load to processor $1$ or processor
$3$. If it sends load to processor $1$ it will not satisfy condition
(\ref{eq:ping-pong}) because after the sending it will be less loaded that
-$x_3^2(t)$. So we consider that the \texttt{ping-pong} condition is probably to
+$x_3^2(t)$. So we consider that the \emph{ping-pong} condition is probably to
strong. Currently, we did not try to make another convergence proof without this
condition or with a weaker condition.
\section{Best effort strategy}
\label{Best-effort}
-\textbf{À traduire} Ordonne les voisins du moins chargé au plus chargé.
-Trouve ensuite, en les prenant dans ce ordre, le nombre maximal de
-voisins tels que tous ont une charge inférieure à la moyenne des
-charges des voisins sélectionnés, et de soi-même.
-
-Les transferts de charge sont ensuite fait en visant cette moyenne pour
-tous les voisins sélectionnés. On envoie une quantité de charge égale
-à (moyenne - charge\_du\_voisin).
-
-~\\\textbf{Question} faut-il décrire les stratégies makhoul et simple ?
+In this section we describe a new load-balancing strategy that we call
+\emph{best effort}. The general idea behind this strategy is that each
+processor, that detects it has more load than some of its neighbors,
+sends some load to the most of its less loaded neighbors, doing its
+best to reach the equilibrium between those neighbors and himself.
+
+More precisely, when a processor $i$ is in its load-balancing phase,
+he proceeds as following.
+\begin{enumerate}
+\item First, the neighbors are sorted in non-decreasing order of their
+ known loads $x^i_j(t)$.
+
+\item Then, this sorted list is traversed in order to find its largest
+ prefix such as the load of each selected neighbor is lesser than:
+ \begin{itemize}
+ \item the processor's own load, and
+ \item the mean of the loads of the selected neighbors and of the
+ processor's load.
+ \end{itemize}
+ Let's call $S_i(t)$ the set of the selected neighbors, and
+ $\bar{x}(t)$ the mean of the loads of the selected neighbors and of
+ the processor load:
+ \begin{equation*}
+ \bar{x}(t) = \frac{1}{\abs{S_i(t)} + 1}
+ \left( x_i(t) + \sum_{j\in S_i(t)} x^i_j(t) \right)
+ \end{equation*}
+ The following properties hold:
+ \begin{equation*}
+ \begin{cases}
+ S_i(t) \subset V(i) \\
+ x^i_j(t) < x_i(t) & \forall j \in S_i(t) \\
+ x^i_j(t) < \bar{x} & \forall j \in S_i(t) \\
+ x^i_j(t) \leq x^i_k(t) & \forall j \in S_i(t), \forall k \in V(i) \setminus S_i(t) \\
+ \bar{x} \leq x_i(t)
+ \end{cases}
+ \end{equation*}
+
+\item Once this selection is completed, processor $i$ sends to each of
+ the selected neighbor $j\in S_i(t)$ an amount of load $s_{ij}(t) =
+ \bar{x} - x^i_j(t)$.
+
+ From the above equations, and notably from the definition of
+ $\bar{x}$, it can easily be verified that:
+ \begin{equation*}
+ \begin{cases}
+ x_i(t) - \sum_{j\in S_i(t)} s_{ij}(t) = \bar{x} \\
+ x^i_j(t) + s_{ij}(t) = \bar{x} & \forall j \in S_i(t)
+ \end{cases}
+ \end{equation*}
+\end{enumerate}
+
+\section{Other strategies}
+\label{Other}
+
+\textbf{Question} faut-il décrire les stratégies makhoul et simple ?
\paragraph{simple} Tentative de respecter simplement les conditions de Bertsekas.
Parmi les voisins moins chargés que soi, on sélectionne :
\section{Virtual load}
\label{Virtual load}
+In this section, we present the concept of \texttt{virtual load}. In order to
+use this concept, load balancing messages must be sent using two different kinds
+of messages: load information messages and load balancing messages. More
+precisely, a node wanting to send a part of its load to one of its neighbors,
+can first send a load information message containing the load it will send and
+then it can send the load balancing message containing data to be transferred.
+Load information message are really short, consequently they will be received
+very quickly. In opposition, load balancing messages are often bigger and thus
+require more time to be transferred.
+
+The concept of \texttt{virtual load} allows a node that received a load
+information message to integrate the load that it will receive later in its load
+(virtually) and consequently send a (real) part of its load to some of its
+neighbors. In fact, a node that receives a load information message knows that
+later it will receive the corresponding load balancing message containing the
+corresponding data. So if this node detects it is too loaded compared to some
+of its neighbors and if it has enough load (real load), then it can send more
+load to some of its neighbors without waiting the reception of the load
+balancing message.
+
+Doing this, we can expect a faster convergence since nodes have a faster
+information of the load they will receive, so they can take in into account.
+
+\textbf{Question} Est ce qu'on donne l'algo avec virtual load?
+
\section{Simulations}
\label{Simulations}
In order to test and validate our approaches, we wrote a simulator
using the SimGrid
-framework~\cite{casanova+legrand+quinson.2008.simgrid}. The process
-model is detailed in the next section (\ref{Sim model}), then the
-results of the simulations are presented in section~\ref{Results}.
+framework~\cite{casanova+legrand+quinson.2008.simgrid}. This
+simulator, which consists of about 2,700 lines of C++, allows to run
+the different load-balancing strategies under various parameters, such
+as the initial distribution of load, the interconnection topology, the
+characteristics of the running platform, etc. Then several metrics
+are issued that permit to compare the strategies.
+
+The simulation model is detailed in the next section (\ref{Sim
+ model}), and the experimental contexts are described in
+section~\ref{Contexts}. Then the results of the simulations are
+presented in section~\ref{Results}.
\subsection{Simulation model}
\label{Sim model}
-\begin{verbatim}
-Communications
-==============
-
-There are two receiving channels per host: control for information
-messages, and data for load transfers.
-
-Process model
-=============
-
-Each process is made of 3 threads: a receiver thread, a computing
-thread, and a load-balancer thread.
-
-* Receiver thread
- ---------------
-
- Loop
- | wait for a message to come, either on data channel, or on ctrl channel
- | push received message in a buffer of received messages
- | -> ctrl messages on the one side
- | -> data messages on the other side
- +-
-
- The loop terminates when a "finalize" message is received on each
- channel.
-
-* Computing thread
- ----------------
-
- Loop
- | if we received some real load, get it (data messages)
- | if there is some real load to send, send it
- | if we own some load, simulate some computing on it
- | sleep a bit if we are looping too fast
- +-
- send CLOSE on data for all neighbors
- wait for CLOSE on data from all neighbors
-
- The loop terminates when process::still_running() returns false.
- (read the source for full details...)
-
-* Load-balancing thread
- ---------------------
-
- Loop
- | call load-balancing algorithm
- | send ctrl messages
- | sleep (min_lb_iter_duration)
- | receive ctrl messages
- +-
- send CLOSE on ctrl for all neighbors
- wait for CLOSE on ctrl from all neighbors
+In the simulation model the processors exchange messages which are of
+two kinds. First, there are \emph{control messages} which only carry
+information that is exchanged between the processors, such as the
+current load, or the virtual load transfers if this option is
+selected. These messages are rather small, and their size is
+constant. Then, there are \emph{data messages} that carry the real
+load transferred between the processors. The size of a data message
+is a function of the amount of load that it carries, and it can be
+pretty large. In order to receive the messages, each processor has
+two receiving channels, one for each kind of messages. Finally, when
+a message is sent or received, this is done by using the non-blocking
+primitives of SimGrid\footnote{That are \texttt{MSG\_task\_isend()},
+ and \texttt{MSG\_task\_irecv()}.}.
+
+During the simulation, each processor concurrently runs three threads:
+a \emph{receiving thread}, a \emph{computing thread}, and a
+\emph{load-balancing thread}, which we will briefly describe now.
+
+\paragraph{Receiving thread} The receiving thread is in charge of
+waiting for messages to come, either on the control channel, or on the
+data channel. Its behavior is sketched by Algorithm~\ref{algo.recv}.
+When a message is received, it is pushed in a buffer of
+received message, to be later consumed by one of the other threads.
+There are two such buffers, one for the control messages, and one for
+the data messages. The buffers are implemented with a lock-free FIFO
+\cite{sutter.2008.writing} to avoid contention between the threads.
+
+\begin{algorithm}
+ \caption{Receiving thread}
+ \label{algo.recv}
+ \KwData{
+ \begin{algodata}
+ \VAR{ctrl\_chan}, \VAR{data\_chan}
+ & communication channels (control and data) \\
+ \VAR{ctrl\_fifo}, \VAR{data\_fifo}
+ & buffers of received messages (control and data) \\
+ \end{algodata}}
+ \While{true}{%
+ wait for a message to be available on either \VAR{ctrl\_chan},
+ or \VAR{data\_chan}\;
+ \If{a message is available on \VAR{ctrl\_chan}}{%
+ get the message from \VAR{ctrl\_chan}, and push it into \VAR{ctrl\_fifo}\;
+ }
+ \If{a message is available on \VAR{data\_chan}}{%
+ get the message from \VAR{data\_chan}, and push it into \VAR{data\_fifo}\;
+ }
+ }
+\end{algorithm}
+
+\paragraph{Computing thread} The computing thread is in charge of the
+real load management. As exposed in Algorithm~\ref{algo.comp}, it
+iteratively runs the following operations:
+\begin{itemize}
+\item if some load was received from the neighbors, get it;
+\item if there is some load to send to the neighbors, send it;
+\item run some computation, whose duration is function of the current
+ load of the processor.
+\end{itemize}
+Practically, after the computation, the computing thread waits for a
+small amount of time if the iterations are looping too fast (for
+example, when the current load is near zero).
+
+\begin{algorithm}
+ \caption{Computing thread}
+ \label{algo.comp}
+ \KwData{
+ \begin{algodata}
+ \VAR{data\_fifo} & buffer of received data messages \\
+ \VAR{real\_load} & current load \\
+ \end{algodata}}
+ \While{true}{%
+ \If{\VAR{data\_fifo} is empty and $\VAR{real\_load} = 0$}{%
+ wait until a message is pushed into \VAR{data\_fifo}\;
+ }
+ \While{\VAR{data\_fifo} is not empty}{%
+ pop a message from \VAR{data\_fifo}\;
+ get the load embedded in the message, and add it to \VAR{real\_load}\;
+ }
+ \ForEach{neighbor $n$}{%
+ \If{there is some amount of load $a$ to send to $n$}{%
+ send $a$ units of load to $n$, and subtract it from \VAR{real\_load}\;
+ }
+ }
+ \If{$\VAR{real\_load} > 0.0$}{
+ simulate some computation, whose duration is function of \VAR{real\_load}\;
+ ensure that the main loop does not iterate too fast\;
+ }
+ }
+\end{algorithm}
+
+\paragraph{Load-balancing thread} The load-balancing thread is in
+charge of running the load-balancing algorithm, and exchange the
+control messages. It iteratively runs the following operations:
+\begin{itemize}
+\item get the control messages that were received from the neighbors;
+\item run the load-balancing algorithm;
+\item send control messages to the neighbors, to inform them of the
+ processor's current load, and possibly of virtual load transfers;
+\item wait a minimum (configurable) amount of time, to avoid to
+ iterate too fast.
+\end{itemize}
- The loop terminates when process::still_running() returns false.
- (read the source for full details...)
-\end{verbatim}
+\begin{algorithm}
+ \caption{Load-balancing}
+ \label{algo.lb}
+ \While{true}{%
+ \While{\VAR{ctrl\_fifo} is not empty}{%
+ pop a message from \VAR{ctrl\_fifo}\;
+ identify the sender of the message,
+ and update the current knowledge of its load\;
+ }
+ run the load-balancing algorithm to make the decision about load transfers\;
+ \ForEach{neighbor $n$}{%
+ send a control messages to $n$\;
+ }
+ ensure that the main loop does not iterate too fast\;
+ }
+\end{algorithm}
+
+\paragraph{}
+For the sake of simplicity, a few details were voluntary omitted from
+these descriptions. For an exhaustive presentation, we refer to the
+actual code that was used for the experiments, and which is
+available at \textbf{FIXME URL}.
+
+\textbf{FIXME: ajouter des détails sur la gestion de la charge virtuelle ?}
+
+\subsection{Experimental contexts}
+\label{Contexts}
+
+\paragraph{Configurations}
+\begin{description}
+\item[\textbf{platforms}] homogeneous (cluster); heterogeneous (subset
+ of Grid5000)
+\item[\textbf{platform size}] platforms with 16, 64, 256, and 1024 nodes
+\item[\textbf{topologies}] line; torus; hypercube
+\item[\textbf{initial load distribution}] initially on a only node;
+ initially on all nodes
+\item[\textbf{comp/comm ratio}] $10/1$, $1/1$, $1/10$
+\end{description}
+
+\paragraph{Algorithms}
+\begin{description}
+\item[\textbf{strategies}] makhoul; besteffort with $k\in \{1,2,4\}$
+\item[\textbf{variants}] with, and without virtual load (bookkeeping)
+\item[\textbf{domain}] real load, and integer load
+\end{description}
+
+\paragraph{Metrics}
+
+\begin{description}
+\item[\textbf{average idle time}]
+\item[\textbf{average convergence date}]
+\item[\textbf{maximum convergence date}]
+\item[\textbf{data transfer amount}] relative to the total data amount
+\end{description}
\subsection{Validation of our approaches}
\label{Results}
%%% ispell-local-dictionary: "american"
%%% End:
-% LocalWords: Raphaël Couturier Arnaud Giersch Abderrahmane Sider
-% LocalWords: Bertsekas Tsitsiklis SimGrid DASUD
+% LocalWords: Raphaël Couturier Arnaud Giersch Abderrahmane Sider Franche ij
+% LocalWords: Bertsekas Tsitsiklis SimGrid DASUD Comté Béjaïa asynchronism ji
+% LocalWords: ik isend irecv Cortés et al chan ctrl fifo
