Laboratoire de l'Informatique et du Parallélisme » XLcloud

Scenarios,Driver,API,API+1 req/s

IPMI cards signed,27,16,20

PDUs signed,26,16,20

IPMI cards unsigned,20,12,16

PDUs unsigned,11,9,13

def api_graph(input_dir, output_dir):

    print("Creating API CPU usage graph...")

    inf = open(input_dir + '/cpu_usage_api.txt', 'r')

    lines = inf.readlines()

    inf.close()

    figure(figsize=(7, 4))

    ax = plt.subplot(111)

    # fig, ax = plt.subplots()

    width = 0.35  # the width of the bars

    first_line = True

    count = 0

    api = []

    apireq = []

    xticks = []

    for line in lines:

        elems = line.split(',')

        if first_line:

            first_line = False

            continue

        xticks.append(elems[0])

        api.append(int(elems[1]))

        apireq.append(int(elems[2]))

    opacity = 0.5

    N = 4

    ind = [float(0),float(2),float(4),float(6)]

    ind2 = [float(0.8),float(2.8),float(4.8),float(6.8)]

    indt = [float(0.8),float(2.8),float(4.8),float(6.8)]

    #print ind + width

    colors = ['r','b','g','y']

    # bars = ax.bar(ind + width / 2, values, width=width, color=colors, alpha=opacity)

    bars_api = ax.barh(ind, api, color='r', alpha=opacity)

    bars_apireq = ax.barh(ind2, apireq, color='b', alpha=opacity)

    plt.yticks(indt, ('IPMI cards\nsigned', 'PDUs\nsigned', 'IPMI cards\nunsigned', 'PDUs\nunsigned'))

    # colors = np.linspace(0, 1, len(bars))

    # for bar in bars:

    # plt.tick_params(

    #     axis='y',          # changes apply to the y-axis

    #     which='both',      # both major and minor ticks are affected

    #     bottom='off',      # ticks along the bottom edge are off

    #     top='off',         # ticks along the top edge are off

    #     labelbottom='off') # labels along the bottom edge are off

    # plt.ylabel('Scenarios')

    plt.xlabel('CPU Usage (\%)')

    # plt.title('CPU Usage of Drivers Under Different Scenarios')

    plt.legend((bars_api, bars_apireq), ['REST API only', 'REST API + 1 req./s'])

    plt.tight_layout()

    #plt.show()

    plt.savefig(output_dir + '/cpu_usage_api.pdf')

    #cpu_graph(args.input_dir, args.output_dir)

    api_graph(args.input_dir, args.output_dir)

We also evaluated the CPU consumption of the REST API data consumer under the scenarios described in Table \ref{tab:parameters_usage}. In addition to these scenarios, two conditions were assessed, namely (i) the REST API working as a consumer requesting data from drivers at a one-second time interval (REST API only); and (ii) the API requesting data at one-second interval and also answering a call every second to provide the collected data to an external system (REST API + 1 req/s). Figure \ref{fig:cpu_usage_consumer} summarises the obtained results. The CPU consumption is in general low, and even when message signing is enabled and the API serves a query, its consumption is below 20\%. The small variation between the scenarios without message signing is caused by the manner ZeroMQ accumulates data on nodes prior to transmission. 

\begin{figure}[!ht]

\center

\includegraphics[width=1.\columnwidth]{figs/cpu_usage_api.pdf}

\caption{API consumer CPU usage under the evaluated scenarios.}

\label{fig:cpu_usage_consumer}

\end{figure}