MaRIONet/60th.html at master · jeremysinger/MaRIONet · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
<!doctype html>
<html lang="en">
<head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <title>MaRIONet : RS4</title>

<link href="bootstrap.min.css" rel="stylesheet">
<link rel="stylesheet" href="style.css">
<link rel="stylesheet" href="https://unpkg.com/purecss@0.6.1/build/pure-min.css">


        <!--[if lte IE 8]>
            <link rel="stylesheet" href="https://purecss.io/combo/1.18.13?/css/layouts/side-menu-old-ie.css">
        <![endif]-->
        <!--[if gt IE 8]><!-->
            <link rel="stylesheet" href="https://purecss.io/combo/1.18.13?/css/layouts/side-menu.css">
        <!--<![endif]-->
    <!--[if lt IE 9]>
        <script src="https://cdnjs.cloudflare.com/ajax/libs/html5shiv/3.7/html5shiv.js"></script>
    <![endif]-->

</head>

<body>

<div id="layout">

   <!-- Menu toggle -->
    <a href="#menu" id="menuLink" class="menu-link">
        <!-- random icon -->
        <span></span>
    </a>

    <div id="menu">
        <div class="pure-menu">
            <a class="pure-menu-heading" href="#">Menu</a>

            <ul class="pure-menu-list">
                <li class="pure-menu-item">
                  <a href="index.html" class="pure-menu-link">Home</a>
                </li>
                <li class="pure-menu-item">
                  <a href="http://gow.epsrc.ac.uk/NGBOViewGrant.aspx?GrantRef=EP/P006434/1" class="pure-menu-link">About</a>
                </li>
                <li class="pure-menu-item" class="menu-item-divided pure-menu-selected">
                  <a href="form.html" class="pure-menu-link">Register</a>
                </li>

				<li class="pure-menu-item" class="menu-item-divided pure-menu-selected">
                  <a href="members.html" class="pure-menu-link">Members</a>
                </li>

				<li class="pure-menu-item" class="menu-item-divided pure-menu-selected">
                  <a href="events.html" class="pure-menu-link">Events</a>
                </li>

				<li class="pure-menu-item" class="menu-item-divided pure-menu-selected">
                  <a href="joining.html" class="pure-menu-link">Joining</a>
                </li>

                <li class="pure-menu-item">
                  <a href="contact.html" class="pure-menu-link">Contact</a>
                </li>
            </ul>
        </div>
    </div>


<div id="main">

<div class="header">
    <img class="pure-img" src="marionet_logo.png" alt="marionet logo">
  </div>


<div class="content">


  <h2 class="content-subhead">RS4</h2>

  <h4>Reliable, Secure and Scalable Software Systems (RS4)</h4>
  <h4>Friday 1st September 2017</h4>
  <h4>University of Glasgow</h4>

  <p>
We had a very productive workshop with 7 talks and some plenary discussion. The workshop was
attended by some 50 people with a good mix of industrial and academic speakers and participants.
There were presentations from high profile industrialists: Alastair Reid (ARM), Alastair Murray
(Codeplay Software), Michele Weiland (EPCC), David Irvine (Maidsafe) and Tony Printezis (Twitter),
and participation from leading universities: including Bristol, Cambridge, Edinburgh, Glasgow, and
Heriot-Watt.
<p>
We discussed the challenges of engineering reliable scalable and secure software (RS4), and some
possible approaches to solving them. Some key outcomes and ongoing activities include:
<p>
(1) Identification of good research ideas and strategies that address modern software and systems
engineering problems;
<p>
(2) Using RS4 as a platform to capture some software and systems engineering success stories and
challenges from industry and academic viewpoints;
<p>
(3) The identification of complementarities and synergies, as well as possibilities for
collaboration/results adoptions between industry and academic participants;
<p>
(4) More broadly, RS4 was aimed at identifying new challenges and trends to influence the UK and
European research agenda, e.g. via the EPSRC MaRIONet and EU HiPEAC Networks. We note that the
topic is closely aligned with the EPSRC’s newly launched “Safe and Secure ICT” priority.
<p>

  <h2>Talks</h2>
  <table class="mq-table pure-table-bordered pure-table">
<tbody>
<tr class="pure-table-odd">
<td> Talk 1 </td>
<td> Alastair Reid - ARM
<br/>Can you trust formally verified software?
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseMarcoCianfriglia" aria-expanded="false" aria-controls="collapseMarcoCianfriglia">
  Abstract
</button>
<div class="collapse abstract" id="collapseMarcoCianfriglia">
A recent, important advance in improving security is that formal verification of system software has
finally started to become viable: we have examples of formally verified microkernels, realistic
compilers, hypervisors etc. These are huge achievements and we can expect to see even more
impressive results in the future but the correctness proofs depend on a number of assumptions
about the Trusted Computing Base that the software depends on. Two key questions to ask are: Are
the specifications of the Trusted Computing Base correct? And do the implementations match the
specifications? I will explore the philosophical challenges and practical steps you can take in
answering that question for one of the major dependencies: the hardware your software runs on. I
will describe the combination of formal verification and testing that ARM uses to verify theprocessor specification and I will talk about our current challenge: getting the specification down to
zero bugs while the architecture continues to evolve. </div>
</td>
</tr>
<tr>
<td class="time" style="width:100px">Talk 2</td>
<td> Jonathan Woodruff - Cambridge Computer Laboratory
<br/>The Security and Scalability of Capability Memory Protection
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseTimLaw" aria-expanded="false" aria-controls="collapseTimLaw">
  Abstract
</button>
<div class="collapse abstract" id="collapseTimLaw">
New, stronger security mechanisms must also scale with system size in order to be useful in today's
ecosystems. Leading security proposals for fine-grained isolation add TLB-like mechanisms that
require system-wide shoot-downs, and others recommend parallel memory hierarchies for tags.
CHERI capability protection enables both fine-grained memory safety and strong intra-program
isolation without adding any scaling challenges to our existing consistency mechanisms. Today's
large systems must devote significant resources to preserve consistency in memory, and CHERI
capabilities exploit the existing memory-consistency mechanisms by placing all protection primitives
into physical memory, with capabilities and their tags atomically joined as 257-bit words. As a result
we have found there to be no significant barrier to adoption to the CHERI model, not only in
software infrastructure including LLVM and FreeBSD, but also for large-scale hardware design.
</div>
</td>
</tr>
<tr class="pure-table-odd">
<td >Talk 3</td>
<td>Tony Printezis - Twitter
<br/>Life of a Twitter JVM engineer
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseChristopherBrown" aria-expanded="false" aria-controls="collapseChristopherBrown">
  Abstract
</button>
<div class="collapse abstract" id="collapseChristopherBrown">
Twitter runs a swarm of services executing on several different managed runtimes. The services have
migrated from mainly Ruby in the past to mostly Scala today. Targeting the JVM allows developers to
quickly write and deploy scalable and reliable code. Automated memory management, in particular,
improves productivity of teams in a fast paced environment. But with these benefits also come
challenges. The way code is executed on the JVM makes trouble-shooting quite tricky, especially in
production, as it is often difficult for developers to understand what has gone wrong and how to fix
it. To make matters even more challenging, the sheer volume of data that flows through these
services, and their stringent latency requirements, stress the JVM in different and novel ways.
The talk will cover:
<br/>
- how services are deployed and monitored at Twitter,<br/>
- why Twitter has a #TwitterVMTeam, and<br/>
- what are the benefits of using a custom JVM and JDK with in-house features.
</div></td>
</tr>
<tr>
<td class="time">Talk 4</td><td>Dimitrios Pezaros - University of Glasgow
<br/>Infrastructure support for future resilient networked systems
<br/><button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseKeynote" aria-expanded="false" aria-controls="collapseKeynote">
  Abstract
</button>
<div class="collapse abstract" id="collapseKeynote">
Recent major cyber-security incidents (e.g., the NHS ransomware incident in 2017) and high-profile
Distributed Denial of Service attacks (e.g., the Mirai botnet attack in 2016) have demonstrated that
the current, context-agnostic (inter-)network architecture is not sufficient to provide for resilient
communications at the onset of adversarial events. The model of the un-intelligent, data-forwarding
network where (security) services are solely provided by isolated end-systems is deemed to always
be one-step behind the most recently-exploited software vulnerabilities. In this talk, we will discuss
work conducted within a recent EPSRC-funded project that tries to introduce the concept of
situational awareness for future, mission-critical networked systems. In particular, we will look athow recent advances in networking technology can be exploited to provide infrastructural support
for the flexible deployment of proactive services such as, e.g., always-on traffic monitoring and
infrastructure protection against attacks. This work constitutes a first step towards future, intelligent
(and resilient) information infrastructures.
</div>
</td>
</tr>
<tr class="pure-table-odd">
<td >Talk 5</td> <td>David Irvine- Maidsafe
<br/>Scalability of data networking
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseHansVan" aria-expanded="false" aria-controls="collapseHansVan">
  Abstract
</button>
<div class="collapse abstract" id="collapseHansVan">
There are many advances in software distribution security, such as secure multi layer signing by
disparate non local parties. This talk will not cover that area, but instead focus on secure data
networking.<br/>
Secure data networking has at it's core a major problem to overcome, should we be successful as a
society. That is secured ownership of the data silo's or some mechanism to achieve trust of the
owning party. In MaidSafe we have taken the approach that the answer to this question is nobody
owns the data silo (network), but everyone owns their own data on it. This sounds confusing and
complex, but in fact it is the most natural mechanism we know from our own history.
Just as we shared written papers between parties in the past and sent these in encrypted form, or
secured behind a seal, MaidSafe has designed a network that follows this natural paradigm. This
means turning the Internet into a connected network of many nodes all provide resource, instead of
the current mechanism of several "trusted" or "capable" entities holding our data within political
borders behind pseudo regulated contracts.<br/>
The MaidSafe solution is an autonomous network of connected computers where the network, and
access to it, belongs to nobody. The Secure Access For Everyone (SAFE) Network is without owners,
requiring no trust between peers, and no intermediary to enable access, just as the free flow of data
is supposed to be.<br/>
I hope this talk will be enlightening and at least leave the group with questions that will challenge
the status quo.
</div></td>
</tr>
<tr>
<td class="time">Talk 6<td>Grzegorz Pawelczak, Simon McIntosh-Smith - University of Bristol
<br/>Application-Based Fault Tolerance Techniques for Sparse Matrix Solvers
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseOracle" aria-expanded="false" aria-controls="collapseOracle">
  Abstract
</button>
<div class="collapse abstract" id="collapseOracle">
The Mean Time Between Failures is expected to decrease as the size and complexity highperformance computing systems continues to increase. With this increase in the number of
components, Fault Tolerance (FT) has been identified as one of the major challenges for exascale
computing. One source of faults in a system are soft errors caused by cosmic rays, which can cause
single or multi bit corruptions to the data held in memory.<br/>
Current solutions for protection against soft errors include hardware Error Correcting Codes (ECC).
These have some disadvantages, including that they consume more power, require extra memory
bandwidth and storage, and they also introduce more complexity to the hardware. ApplicationBased Fault Tolerance (ABFT) allows us to adapt an Error Detection and Correction technique to the
application and its data structure.In this presentation we demonstrate ABFT techniques for Sparse Matrix Solvers using TeaLeaf, a heat
conducting miniapp from the Mantevo Project.<br/>
This software-based approach fully protects the sparse matrix stored in the compressed sparse row
format and the dense floating point vectors from soft errors. To protect the data we compare
different error detection and/or correction methods such as Hamming Codes and CRC, and identify
the trade-offs between them. Our solution requires no extra memory storage as the redundant data
for the error detection and correction is stored in the least significant bits of the mantissa or in the
unused bits from the integer vectors.<br/>
We also investigate tradeoff between the error detection/correction interval and the accuracy of the
linear solve - performing the test and correction only once per `n’ timesteps can significantly reduce
the performance overhead of the approach, at the cost of potentially performing more redundant
work after an error has occurred but before it is detected.<br/>
In this presentation we will explain the details of these Application-Based Fault Tolerance techniques
and how they combat soft errors, as well as presenting performance results on different
architectures including x86, ARM, Intel Xeon Phi (Knights Landing) and GPUs.
</div></td>
</tr>
<tr class="pure-table-odd">
<td class="time">Talk 7</td> <td>Alastair Murray - Codeplay Software
<br/>Implementing Heterogeneous Languages for Safety Critical Use-Cases
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapsePabloGonzales" aria-expanded="false" aria-controls="collapsePabloGonzales">
  Abstract
</button>
<div class="collapse abstract" id="collapsePabloGonzales">
Until recently safety critical programmers have largely avoided parallel hardware. However, great
advances in machine vision and deep learning have been made by exploiting the computational
power of parallel devices such as GPUs or DSPs. These new technologies are now being used in
safety critical contexts, for example self-driving cars, meaning that the heterogeneous language
runtimes must be both reliable and predictable. This talk will discuss changes required to current
parallel programming languages such as OpenCL so that they are both usable and implementable in
a safety critical context, and some initial thoughts on actually implementing a safe heterogeneous
language runtime.
</div></td>
</tr>
<tr>
<td class="time">Talk 8</td> <td>Anna Lito Michala, Jeremy Singer -University of Glasgow
<br/>Scaling for the Future: AnyScale Apps
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseJoseNunez" aria-expanded="false" aria-controls="collapseJoseNunez">
  Abstract
</button>
<div class="collapse abstract" id="collapseJoseNunez">
The Anyscale Apps project involves the development of tools that will enable the creation of
software capable of running on any scale of device, from IoT wireless systems to large data centres.
The project is funded by EPSRC and aims to cater for a heterogeneous future of versatile
infrastructure. Automatic adapation of software to different underlying hardware will enable faster
development cycles and support the next generation of computing systems. Two demonstrators
were identified as case studies to test our tools.<br/>
The smart building demonstrator incorporates a network of commodity low cost hardware including
sensors to monitor occupancy of meeting rooms in the University of Glasgow School of Computing
Science. In this case, edge computing is utilised to exploit the computational capabilities of the
underlying hardware while being low cost and easy to scale based on a modular software design.The second case is a robot demonstrator. The distribution of load is performed through a pack of
robots in order to identify a specific person in a busy room. The robots are required to perform
compute-intensive tasks on low capability hardware and thus rely on offloading to perform the
required task within acceptable time frames.<br/>
Both these demonstrators identify the limitations of the developed tools as well as their advantages
and usability.
</div></td>
</tr>
<tr class="pure-table-odd">
<td class="time">Talk 9</td> <td>Michele Weiland - EPCC
<br/>Automatically Deriving Cost Models for Structured Parallel Processes Using Hylomorphisms
<br/>
<button class="my-btn" type="button" data-toggle="collapse" data-target="#collapseKevinHammond" aria-expanded="false" aria-controls="collapseKevinHammond">
  Abstract
</button>
<div class="collapse abstract" id="collapseKevinHammond">
This talk will give an overview of the different aspects of the Exascale challenge and assess the
progress towards reaching this goal. EPCC, the supercomputing centre at the University of
Edinburgh, is involved (and leading) a variety of efforts in the Exascale research programme, some of
which will be presented here.
</div></td>
</tr>
</tbody>
</table>


</div> <!-- content -->

</div> <!-- main -->
</div>


 <!-- jQuery -->
<script src="js/jquery.js"></script>
 <!-- Bootstrap Core JavaScript -->
<script src="js/bootstrap.min.js"></script>

    <!-- Custom Theme JavaScript -->
    <script>
    // Closes the sidebar menu
    $("#menu-close").click(function(e) {
        e.preventDefault();
        $("#sidebar-wrapper").toggleClass("active");
    });

    // Opens the sidebar menu
    $("#menu-toggle").click(function(e) {
        e.preventDefault();
        $("#sidebar-wrapper").toggleClass("active");
    });

    // Scrolls to the selected menu item on the page
    $(function() {
        $('a[href*=#]:not([href=#])').click(function() {
            if (location.pathname.replace(/^\//, '') == this.pathname.replace(/^\//, '') || location.hostname == this.hostname) {

                var target = $(this.hash);
                target = target.length ? target : $('[name=' + this.hash.slice(1) + ']');
                if (target.length) {
                    $('html,body').animate({
                        scrollTop: target.offset().top
                    }, 1000);
                    return false;
                }
            }
        });
    });
    </script>

<script src="https://purecss.io/combo/1.18.13?/js/ui.js"></script>

</body>
</html>