New bandwidth curve
This commit is contained in:
@@ -300,6 +300,54 @@ These tables also provide information about the simulation time of the different
|
||||
\label{fig:latency_ddr4}
|
||||
\end{figure}
|
||||
|
||||
In order to compare not only the total average bandwidth, but also its behavior over time, all benchmarks were run consecutively on gem5 SE amd DRAMSys and plotted as a bandwidth-time diagram in Figure \ref{fig:data_bus_util}.
|
||||
|
||||
\begin{figure}[!ht]
|
||||
\begin{center}
|
||||
\begin{tikzpicture}
|
||||
\begin{axis}[
|
||||
width=\linewidth,
|
||||
xlabel={Relative Simulation Time},
|
||||
ylabel={Data Bus Utilization},
|
||||
xmin=0, xmax=1,
|
||||
ymin=0,
|
||||
ytick={0,0.1,0.2,0.3},
|
||||
legend style={
|
||||
at={(current bounding box.south-|current axis.south)},
|
||||
anchor=north,
|
||||
legend columns=-1,
|
||||
draw=none,
|
||||
/tikz/every even column/.append style={column sep=0.5cm}
|
||||
},
|
||||
ymajorgrids=true,
|
||||
grid style=dashed,
|
||||
]
|
||||
|
||||
\addplot[mark=none, color=blue] table [x=T,y=B,col sep=comma] {dat/gem5.csv};
|
||||
|
||||
\addplot[mark=none, color=red] table [x=T,y=B,col sep=comma] {dat/dramsys.csv};
|
||||
|
||||
\legend{gem5 SE, DRAMSys}
|
||||
|
||||
\end{axis}
|
||||
\end{tikzpicture}
|
||||
\end{center}
|
||||
\caption{Data bus utilization over the simulation time with DDR4-2400.}
|
||||
\label{fig:data_bus_util}
|
||||
\end{figure}
|
||||
|
||||
Similar to the previous comparisons, the bandwidth on average of DRAMSys is marginally lower than gem5.
|
||||
Furthermore, an increased fluctuation around a bandwidth value can be observed.
|
||||
However, the overall time behavior is the same:
|
||||
The highest bandwidth value is reached at the beginning of the simulation and the bandwidth drops to a low plateau at the end of the simulation.
|
||||
|
||||
The \revabbr{mean absolute percentage error}{MAPE} is used to evaluate the deviations of the two bandwidth curves.
|
||||
With the total number of data points $n$, the individual actual values $A_t$ and the individual forecasted values $F_t$, the measure is defined as follows:
|
||||
\begin{equation}
|
||||
MAPE=\frac{1}{n}\sum_{t=1}^{n}\bigg|\frac{A_t-F_t}{A_t}\bigg|
|
||||
\end{equation}
|
||||
For the bandwidth simulations, the resulting MAPE value is 22.3\%.
|
||||
|
||||
\subsection{Comparison to Ramulator}
|
||||
|
||||
In order to evaluate the new simulation frontend with a simulator that uses a similar approach, the benchmarks are compared with Ramulator in this section.
|
||||
|
||||
Reference in New Issue
Block a user