[Ikeda, Yasui-san] 概要 DAQ Timig Errorの調査の続きです。 K-Log#20464の調査でHyperThreadはDisableの方が全体の効率は上がる可能性を試した。 https://klog.icrr.u-tokyo.ac.jp/osl/?r=20464 しかし現状あまり効果が分からない。 少し古い情報かもしれないがT1500227の5.3.2にはHyperthreading - Enabled(DAQ code is multi-threaded)との記載がある。 https://dcc.ligo.org/public/0119/T1500227/001/LIGO-T1500227-v1.pdf 対応 k1dc0のHyperThread設定をBIOSでEnableに変更しました。(Fig.1) 次に前回K-Log#20563にてDAQエラーが発生した方法を試した。 https://klog.icrr.u-tokyo.ac.jp/osl/?r=20563 > 2. Connected all RTPCs (C1, C2, C3) directly to the SW of NETGEAR to which k1dc0 is connected. k1dc0と同じSWに直接接続したがエラーの発生頻度が低く改善していた。 これらの変更前は、数回/日の発生頻度であった。 今週はこのまま放置します。 --- [Ikeda, Yasui-san] Summary This is a continuation of the DAQ Timig Error investigation. In the investigation of K-Log#20464, I tested the possibility that the overall efficiency might be higher if HyperThread is Disable. https://klog.icrr.u-tokyo.ac.jp/osl/?r=20464 However, I am not sure of the effect at present. This may be a bit out of date, but T1500227 5.3.2 states that Hyperthreading - Enabled (DAQ code is multi-threaded). https://dcc.ligo.org/public/0119/T1500227/001/LIGO-T1500227-v1.pdf Correspondence We changed the HyperThread setting of k1dc0 to Enable in the BIOS.(Fig.1) Next, We tried the method that caused the DAQ error in K-Log#20563 last time. https://klog.icrr.u-tokyo.ac.jp/osl/?r=20563 > 2. Connected all RTPCs (C1, C2, C3) directly to the SW of NETGEAR to which k1dc0 is connected. Connected directly to the same SW as k1dc0 but the error frequency was low and improved. Before these changes, the frequency of occurrence was several times/day. We will leave it as is for the rest of the week.