sirius-db.github.io/index.html at main · sirius-db/sirius-db.github.io · 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
<!doctype html>
<html lang="en">
  <head>
  <meta charset="utf-8">
  <meta http-equiv="x-ua-compatible" content="ie=edge">
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <link rel="preload" href="https://sirius-db.com/fonts/vendor/jost/jost-v4-latin-regular.woff2" as="font" type="font/woff2" crossorigin>
<link rel="preload" href="https://sirius-db.com/fonts/vendor/jost/jost-v4-latin-500.woff2" as="font" type="font/woff2" crossorigin>
<link rel="preload" href="https://sirius-db.com/fonts/vendor/jost/jost-v4-latin-700.woff2" as="font" type="font/woff2" crossorigin>


<link rel="stylesheet" href="/main.26a4124948d1120d25e24277deece65f46efb67c7bb053609c6052e2c04ab8cff459e6b9e42253d24c5ac1deb374c292717ce5486b1d29ccdde8f4bd3bdccb44.css" integrity="sha512-JqQSSUjREg0l4kJ33uzmX0bvtnx7sFNgnGBS4sBKuM/0Wea55CJT0kxawd6zdMKScXzlSGsdKczd6PS9O9zLRA==" crossorigin="anonymous">

<noscript><style>img.lazyload { display: none; }</style></noscript>
  <base href="https://sirius-db.com/">

  <link rel="canonical" href="sirius-db.com">

  <link rel="alternate" type="application/rss+xml" href="https://sirius-db.com/index.xml" title="Sirius">

<title>Sirius</title>
<meta name="description" content="Sirius is a GPU acceleration layer for DuckDB and other SQL analytics engines, delivering order-of-magnitude faster analytics by fully unleashing the power of modern GPUs.">


    <link rel="icon" href="/favicon.ico" sizes="32x32">

      <link rel="icon" href="/favicon.svg" type="image/svg+xml">

      <link
        rel="apple-touch-icon"
        href="/apple-touch-icon.png"
        sizes="180x180"
        type="image/png"
      >
      <link
        rel="icon"
        href="/favicon-192x192.png"
        sizes="192x192"
        type="image/png"
      >
      <link
        rel="icon"
        href="/favicon-512x512.png"
        sizes="512x512"
        type="image/png"
      >
<link rel="manifest" href="/manifest.webmanifest">

<meta property="og:url" content="https://sirius-db.com/">
  <meta property="og:site_name" content="Sirius">
  <meta property="og:title" content="Sirius">
  <meta property="og:description" content="Sirius is a GPU-native SQL engine to accelearate DuckDB and other SQL databases, delivering order-of-magnitude faster analytics by fully unleashing the power of modern GPUs.">
  <meta property="og:locale" content="en">
  <meta property="og:type" content="website">
    <meta property="og:image" content="https://sirius-db.com/favicon.png">

  <meta name="twitter:card" content="summary_large_image">
  <meta name="twitter:image" content="https://sirius-db.com/favicon.png">
  <meta name="twitter:title" content="Sirius">
  <meta name="twitter:description" content="Sirius is a GPU-native SQL engine to accelearate DuckDB and other SQL databases, delivering order-of-magnitude faster analytics by fully unleashing the power of modern GPUs.">
      <meta name="twitter:site" content="@getdoks">
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "WebSite",
      "url": "https://sirius-db.com/",
      "name": "Sirius",
      "description": "Sirius is a GPU-native SQL engine to accelearate DuckDB and other SQL databases, delivering order-of-magnitude faster analytics by fully unleashing the power of modern GPUs."
    }
  ]
}
</script>

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Organization",
      "url": "https://sirius-db.com/",
      "name": "Sirius",
      "image": [
        {
          "@id": "https://sirius-db.com/sirius-logo.png"
        }
      ],
      "logo": {
        "@id": "https://sirius-db.com/images/favicon.png"
      }
    }
  ]
}
</script>


<script async src="https://www.googletagmanager.com/gtag/js?id=G-60DHFLN477"></script>
<script>
  window.dataLayer = window.dataLayer || [];
  function gtag(){dataLayer.push(arguments);}
  gtag('js', new Date());

  gtag('config', 'G-60DHFLN477');
</script>
  </head>


  <body class="single home">

<section class="section container-fluid mt-n3 pb-3">

  <div class="row align-items-center mb-5">
    <div class="col-md-8 text-center text-md-start">
      <div class="d-flex align-items-center justify-content-center justify-content-md-center w-100">
        <img src="/images/sirius-logo.png" alt="Sirius Logo" class="img-fluid me-3" style="max-height: 140px;">
        <h1 class="mb-0" style="font-size: 4.5rem;">Sirius</h1>
      </div>
      <p class="lead fs-3 mb-4 d-flex align-items-center justify-content-center justify-content-md-center w-100">SQL Analytics at GPU Speed</p>
    </div>
    <div class="col-md-8 text-center text-md-start mt-4" >
      <p class="fs-4"><strong>Sirius is a GPU-native SQL engine</strong>. It plugs into existing databases such as DuckDB via the standard Substrait query format, requiring no query rewrites or major system changes. By offloading query execution to GPUs, Sirius achieves over 10× speedup at the same hardware rental cost!
      </p>
    </div>
  </div>
  <div class="d-flex justify-content-start mb-5">
    <div class="bg-dots"></div>
  </div>
  <div class="d-flex flex-column flex-md-row md-align-items-start justify-content-center gap-4 my-2">
    <div class="d-flex flex-column align-items-center col-md-4">
      <a href="mailto:siriusdb@cs.wisc.edu"
         class="btn btn-primary w-40 mb-2"
         style="width: 160px; font-size: 1.5rem; font-weight: 600;">
        <strong>Email Us</strong>
      </a>
      <div class="text-center">We’d love to hear from you — questions, contributions, or collaborations.</div>
    </div>

    <div class="d-flex flex-column align-items-center col-md-4">
      <a href="https://join.slack.com/t/sirius-db/shared_invite/zt-33tuwt1sk-aa2dk0EU_dNjklSjIGW3vg"
         class="btn btn-warning w-80 mb-2"
         style="width: 160px; font-size: 1.5rem; font-weight: 600;">
        <strong>Join Slack</strong>
      </a>
      <div class="text-center">Connect with the community, ask questions, and stay updated.</div>
    </div>

    <div class="d-flex flex-column align-items-center col-md-4">
      <a class="btn btn-dark w-80 mb-2"
         href="https://github.com/sirius-db/sirius"
         target="_blank"
         style="width: 160px; font-size: 1.5rem; font-weight: 600;" >
        <strong>GitHub</strong>
      </a>
      <div class="text-center">Source code available on GitHub</div>
    </div>

    </div>
  </div>


</section>

<section class="section section-md section-features">
    <div class="container">
      <div class="row justify-content-center text-center">
        <h2 class="mb-4">Key Features</h2>
        <div class="col-lg-5">
          <h2 class="h3">GPU-Native Speed</h2>
          <p>Sirius is <strong>10× faster than DuckDB and 60× faster than ClickHouse at the same hardware rental cost</strong>. Built for GPUs, Sirius targets 100× speedups over CPU-based SQL engines.</p>
        </div>
        <div class="col-lg-5">
          <h2 class="h3">Seamless Integration</h2>
          <p>Sirius plugs into databases like DuckDB and Doris via <strong>Substrait</strong>, accelerating SQL workflows without changing your stack. It supports <strong>CPU fallback</strong> for full compatibility.</p>
        </div>
        <div class="col-lg-5">
          <h2 class="h3">Deploy Anywhere</h2>
          <p>Sirius supports a wide range of deployment options, <strong>from cloud to on-prem</strong>, delivering GPU-accelerated performance wherever you run.</p>
        </div>
      </div>
    </div>
  </section>


<section class="section section-md section-arch">
  <div class="row justify-content-center">
    <div class="col-md-8 text-center">
      <h2 class=>Architecture</h2>
      <img src="/images/super-sirius-arch.png" alt="Sirius Architecture Diagram" class="img-fluid">
    </div>
  </div>
</section>


<section class="section section-md section-pub">
  <div class="row justify-content-center mb-5">
    <div class="col-md-10">
      <h2 class="mb-4 text-center">Publications</h2>
      <ul class="list-group list-group-flush">

        <li class="list-group-item">

            <strong><a href="https://vldb.org/cidrdb/papers/2026/p12-yogatama.pdf" target="_blank" rel="noopener noreferrer">Rethinking Analytical Processing in the GPU Era</a></strong><br>

          <span>Bobbi Yogatama*, Yifei Yang*, Kevin Kristensen, Devesh Sarda, Abigale Kim, Adrian Cockcroft, Yu Teng, Joshua Patterson, Gregory Kimball, Wes McKinney, Weiwei Gong, Xiangyao Yu</span><br>
          <strong>CIDR</strong>, 2026<br>

            * Equal Contribution

        </li>

        <li class="list-group-item">

            <strong><a href="https://arxiv.org/pdf/2508.05029" target="_blank" rel="noopener noreferrer">Theseus: A Distributed and Scalable GPU-Accelerated Query Processing Platform Optimized for Efficient Data Movement</a></strong><br>

          <span>Felipe Aramburu, William Malpica, Kaouther Abrougui, Amin Aramoon, Romulo Auccapuclla, Claude Brisson, Matthijs Brobbel, Colby Farrell, Pradeep Garigipati, Joost Hoozemans, Supun Kamburugamuve, Akhil Nair, Alexander Ocsa, Johan Peltenburg, Ruben Quesada Lopez, Deepak Sihag, Ahmet Uyar, Dhruv Vats, Michael Wendt, Jignesh M. Patel, Rodrigo Aramburu</span><br>
          <strong>arXiv</strong>, 2025<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://dl.acm.org/doi/10.1145/3725283" target="_blank" rel="noopener noreferrer">Debunking the Myth of Join Ordering: Toward Robust SQL Analytics</a></strong><br>

          <span>Junyi Zhao, Kai Su, Yifei Yang, Xiangyao Yu, Paraschos Koutris, Huanchen Zhang</span><br>
          <strong>SIGMOD</strong>, 2025<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://dl.acm.org/doi/10.1145/3725259" target="_blank" rel="noopener noreferrer">Accelerate Distributed Joins with Predicate Transfer</a></strong><br>

          <span>Yifei Yang, Xiangyao Yu</span><br>
          <strong>SIGMOD</strong>, 2025<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://pages.cs.wisc.edu/~yxy/pubs/sirius-hpts24.pdf" target="_blank" rel="noopener noreferrer">GPU Databases—The New Modality of Database Analytics (presentation)</a></strong><br>

          <span>Bobbi Yogatama, Xiangyao Yu</span><br>
          <strong>HPTS</strong>, 2024<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://www.vldb.org/pvldb/vol17/p4709-yogatama.pdf" target="_blank" rel="noopener noreferrer">Scaling your Hybrid CPU-GPU DBMS to Multiple GPUs</a></strong><br>

          <span>Bobbi Yogatama, Weiwei Gong, Xiangyao Yu</span><br>
          <strong>VLDB</strong>, 2024<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://www.cidrdb.org/cidr2024/papers/p22-yang.pdf" target="_blank" rel="noopener noreferrer">Predicate Transfer: Efficient Pre-Filtering on Multi-Join Queries</a></strong><br>

          <span>Yifei Yang, Hangdong Zhao, Xiangyao Yu, Paraschos Koutris</span><br>
          <strong>CIDR</strong>, 2024<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://dl.acm.org/doi/10.1145/3592980.3595307" target="_blank" rel="noopener noreferrer">Accelerating User-Defined Aggregate Function (UDAF) with Block-wide Execution and JIT Compilation on GPUs</a></strong><br>

          <span>Bobbi Yogatama , Brandon Miller, Yunsong Wang, Graham Markall, Jake Hemstad, Gregory Kimball, Xiangyao Yu</span><br>
          <strong>DaMoN@SIGMOD</strong>, 2023<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://www.vldb.org/pvldb/vol15/p2491-yogatama.pdf" target="_blank" rel="noopener noreferrer">Orchestrating Data Placement and Query Execution in Heterogeneous CPU-GPU DBMS</a></strong><br>

          <span>Bobbi Yogatama, Weiwei Gong, Xiangyao Yu</span><br>
          <strong>VLDB</strong>, 2022<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://dl.acm.org/doi/10.1145/3514221.3526132" target="_blank" rel="noopener noreferrer">Tile-based Lightweight Integer Compression in GPU</a></strong><br>

          <span>Anil Shanbhag*, Bobbi Yogatama*, Xiangyao Yu, Samuel Madden</span><br>
          <strong>SIGMOD</strong>, 2022<br>

        </li>

        <li class="list-group-item">

            <strong><a href="https://dl.acm.org/doi/abs/10.1145/3318464.3380595" target="_blank" rel="noopener noreferrer">A Study of the Fundamental Performance Characteristics of GPUs and CPUs for Database Analytics</a></strong><br>

          <span>Anil Shanbhag, Samuel Madden, Xiangyao Yu</span><br>
          <strong>SIGMOD</strong>, 2020<br>

        </li>

      </ul>
    </div>
  </div>
</section>


<script async
src="/js/app.1387e6996ce3ad4055b08b809a74544dd1f9c902d2ba87b00850edd5b5f0daeb.js"
integrity="sha256-E4fmmWzjrUBVsIuAmnRUTdH5yQLSuoewCFDt1bXw2us="></script>


    </body>
</html>