From 5fa47448ad6a76f15500e8236915727faa39cd69 Mon Sep 17 00:00:00 2001 From: Stevan Andjelkovic Date: Wed, 17 Nov 2021 08:27:33 +0100 Subject: [PATCH] docs: make some old slides more context free --- doc/demo/slides.md | 2 +- doc/demo/slides.pdf | Bin 313958 -> 313940 bytes src/runtime-prototype/event-loop.org | 6 +++--- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/doc/demo/slides.md b/doc/demo/slides.md index 4d537fe5..1e476231 100644 --- a/doc/demo/slides.md +++ b/doc/demo/slides.md @@ -266,7 +266,7 @@ type fault * Future work * Test case generation, including extending existing/already run test cases; * Support for deployment, for testing upgrades and backup/restore; - * Integrate other parts of Assembly. + * Integrate other parts of platform. --- diff --git a/doc/demo/slides.pdf b/doc/demo/slides.pdf index 4edc1e11c7b8bf25bfec518c7682951e85bf97dc..8c72616502bf10f136962708a5ad0a1692f8282e 100644 GIT binary patch delta 1190 zcmV;X1X=s$loQmH6M%#PgaWh!g%<%Zm+=e(DSyqBO>f&U42JLd6}-&`YFVGQjNJw- z!*&?3)VE>>FP2lIsp8o(y8in~mEtC8x3)=(?GQ_(P4OZ1G8k5X;nNHsti-Owu0O2@ z1Is9qOaRR(A~b_uK{NaIAn2d1hzJM|K?U?HICL!@i^?Q~hm<~O&$D<7QVRZ;`-J!A z;(w{PAIe!erxXZBjAsam5k%Xaw zLqZVc*yS8RlIDd&vx=ryHwEHkiMlo|}ybwNap;o;bhIXSp zpizSv-tQmw60&5<;n>KLa6H6mWAQ38(cQG>((leyrQbR6Fh1!-TAiR{o4a#?2@@z6 z(xOnrFhBC{LxbmzYb$D3g41(=eO)1(*z7WvEV8%nF_+a4fLBT@e+OsY(dy-Ttbbd* z@}||c(p!^TvmK1J(kdp=x>r?w>tFPvc4U?&Ap{j72-obTd_C!%HWue&*Tu;st3cRz z%hyw@ZhdFpUMp?R9(>_vY3j?^StZxXHYKRm{pIcUBV_(jEw9EdwL@wAS?YhA#-NLh zxm2OhuJT<9m2H~fURNeG$Z`o2Z5y{sx z5#I#uP!W;B`hSX0f)*%)Fq}lK|ASB|6jY=V#|#W|KGlFNu$8-3D5~h&7%o8zf0P*+ zq^dwt$r~}!2CSA?E!Ci^EJd{$I-pY}Eyi6p^g@?P>*u(uLyu-FU*b3beVVOy#F70H zo2Gu)4uhJl%|7V z2j}4m%&4`0kM2dd1ef6^Tve-njP5nK4maQ~+)}GeM)x+{fmwJ6b87Y1(Y*)v;Q>5_ zM{2zri_gPjc*5fR8RySn1oG$P z&m(`N(-ZVT6-bzqFze2kuOjIP%hk=whDRENb zWXMN>fH(nh0^-ES$AN4(*>JMqB*!P9srk3BqWSV0`3KTQP(?9YsYu2Bg;+=onfam!8pCWtCvSmf`+o!zQ!gzXv zESt{PwKop-re4o@Rk5u{?|xm-${n7!zD_K;!MD-nmPq}Qe>|;H8a4ASHcy>7b>`RQ z9|Ht9)q{>42= z_^{>V-4lH#=rlD5%BfbXazFH{doxA+LaK^)&$josW=YgPWKnx*c2qdydtzHpaN@;9 zI>&ajuBtZWIlIf%VJ(~Op8NK1Oj8a{IlO73@sig=FFN0CyBNO1I^6a5xh{8? z$$m>Na$0WNX)P|%F=NR_6OB0>QnHob*BvZAy;XFp?!i|#5AItl;FnYO&imoa+wNyl z4J{r-ZRL9ZfJfGMcT9b6>g8)0Qn{;N&U6u#808rLV%uzddu?`s>X>aSQL@_+#fZc7`m|^xOJP zKYzQt-sjYfd*pUs;8}WX^OgT*y?)}^wM%0b#x}+lODXZeXCU$))d` zpW>2OlB%HLVue-7_Sju4WxRn-rmjv#X0C>Au8yXzhRy~CuFi&zmM*S_7OrM)mTty& z3N{3l#B%BTq~@iUWGI+e7(x=Q>ZP6jQxXM^f1Fu-)*vS>jvYC=X_FqR^<Z|pOtHs|PtgYbUzgkuA@?FrTOGgD6pI3n7XXuR9kKub diff --git a/src/runtime-prototype/event-loop.org b/src/runtime-prototype/event-loop.org index 3a0a3cec..fbc40d99 100644 --- a/src/runtime-prototype/event-loop.org +++ b/src/runtime-prototype/event-loop.org @@ -129,7 +129,7 @@ queue | __ __ __ | -. *** scheduler/executor event logging is synchronous and slow *** scheduler implementation is complicated ** Convenience -*** useful for non-async parts of smartlog +*** useful for non-async parts of SUT *** sync communication between reactors that are "near" *** async disk I/O? **** leveldb async flag means no fsync... @@ -182,7 +182,7 @@ queue | __ __ __ | -. * Next steps ** reimplement scheduler on top of event loop -*** some experience using this style of programming before adopting it for Smartlog +*** some experience using this style of programming before adopting it for SUT *** solves problem with slow sync disk I/O *** and slow communication between executor and scheduler *** reduce scheduler complexity, make it testable and debuggable using the detsys @@ -193,7 +193,7 @@ queue | __ __ __ | -. *** same approach as Chuck's event loop, but for client side * Longer term -** put the Smartlog reactors on top of the event loop +** put the SUT reactors on top of the event loop *** timers? ** merge efforts with progress made by others on: *** scalability / benchmarking