Query Performance Tuning Guide

<h3 id="overview" class="position-relative d-flex align-items-center group"> Overview <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="overview" aria-haspopup="dialog" aria-label="Share link: Overview"> Share link </button> </h3><div id="headingShareModal" class="heading-share-modal" role="dialog" aria-modal="true" aria-labelledby="headingShareTitle" hidden> <div class="hsm-dialog" role="document"> <div class="hsm-header"> <h2 id="headingShareTitle" class="h6 mb-0 fw-bold">Share this section</h2> <button type="button" class="hsm-close" aria-label="Close"> </button> </div> <div class="hsm-body"> <label for="headingShareInput" class="form-label small text-muted mb-1 text-uppercase fw-bold" style="font-size: 0.7rem; letter-spacing: 0.5px;">Permalink</label> <div class="input-group mb-4 hsm-url-group"> <input id="headingShareInput" type="text" class="form-control font-monospace" readonly aria-readonly="true" style="font-size: 0.85rem;" /> <button class="btn btn-primary hsm-copy" type="button" aria-label="Copy" title="Copy"> </button> </div> <div class="small fw-bold mb-2 text-muted text-uppercase" style="font-size: 0.7rem; letter-spacing: 0.5px;">Share via</div> <div class="hsm-share-grid"> <a id="share-twitter" class="btn btn-outline-secondary w-100" target="_blank" rel="noopener noreferrer"> Twitter </a> <a id="share-linkedin" class="btn btn-outline-secondary w-100" target="_blank" rel="noopener noreferrer"> LinkedIn </a> <a id="share-facebook" class="btn btn-outline-secondary w-100" target="_blank" rel="noopener noreferrer"> Facebook </a> </div> </div> </div> </div> <style> .heading-share-modal { position: fixed; inset: 0; display: flex; justify-content: center; align-items: center; background: rgba(0, 0, 0, 0.6); z-index: 1050; padding: 1rem; backdrop-filter: blur(4px); -webkit-backdrop-filter: blur(4px); } .heading-share-modal[hidden] { display: none !important; } .hsm-dialog { max-width: 420px; width: 100%; background: var(--bs-body-bg, #fff); color: var(--bs-body-color, #212529); border: 1px solid var(--bs-border-color, rgba(0,0,0,0.1)); border-radius: 1rem; box-shadow: 0 25px 50px -12px rgba(0, 0, 0, 0.25); overflow: hidden; animation: hsm-fade-in 0.2s ease-out; } @keyframes hsm-fade-in { from { opacity: 0; transform: scale(0.95); } to { opacity: 1; transform: scale(1); } } [data-bs-theme="dark"] .hsm-dialog { background: #1e293b; border-color: rgba(255,255,255,0.1); color: #f8f9fa; } .hsm-header { display: flex; justify-content: space-between; align-items: center; padding: 1rem 1.5rem; border-bottom: 1px solid var(--bs-border-color, rgba(0,0,0,0.1)); background: rgba(0,0,0,0.02); } [data-bs-theme="dark"] .hsm-header { background: rgba(255,255,255,0.02); border-color: rgba(255,255,255,0.1); } .hsm-close { background: transparent; border: none; color: inherit; opacity: 0.5; padding: 0.25rem 0.5rem; border-radius: 0.25rem; font-size: 1.2rem; line-height: 1; transition: opacity 0.2s; } .hsm-close:hover { opacity: 1; } .hsm-body { padding: 1.5rem; } .hsm-url-group { display: flex !important; align-items: stretch; } .hsm-url-group .form-control { flex: 1; min-width: 0; margin: 0; background: var(--bs-secondary-bg, #f8f9fa); border-color: var(--bs-border-color, #dee2e6); border-top-right-radius: 0; border-bottom-right-radius: 0; height: 42px; } .hsm-url-group .btn { flex: 0 0 auto; margin: 0; margin-left: -1px; border-top-left-radius: 0; border-bottom-left-radius: 0; height: 42px; display: flex; align-items: center; justify-content: center; padding: 0 1.25rem; z-index: 2; } [data-bs-theme="dark"] .hsm-url-group .form-control { background: #0f172a; border-color: #334155; color: #e2e8f0; } .hsm-share-grid { display: flex; flex-direction: column; gap: 0.5rem; } .hsm-share-grid .btn { display: flex; align-items: center; justify-content: center; font-size: 0.9rem; padding: 0.6rem; border-color: var(--bs-border-color); width: 100%; } [data-bs-theme="dark"] .hsm-share-grid .btn { color: #e2e8f0; border-color: #475569; } [data-bs-theme="dark"] .hsm-share-grid .btn:hover { background: #334155; border-color: #cbd5e1; } </style> <script> (function(){ const modal = document.getElementById('headingShareModal'); if(!modal) return; const input = modal.querySelector('#headingShareInput'); const copyBtn = modal.querySelector('.hsm-copy'); const twitter = modal.querySelector('#share-twitter'); const linkedin = modal.querySelector('#share-linkedin'); const facebook = modal.querySelector('#share-facebook'); const closeBtn = modal.querySelector('.hsm-close'); let lastFocus=null; let trapBound=false; function buildUrl(id){ return window.location.origin + window.location.pathname + '#' + id; } function isOpen(){ return !modal.hasAttribute('hidden'); } function hydrate(id){ const url=buildUrl(id); input.value=url; const enc=encodeURIComponent(url); const text=encodeURIComponent(document.title); if(twitter) twitter.href=`https://twitter.com/intent/tweet?url=${enc}&text=${text}`; if(linkedin) linkedin.href=`https://www.linkedin.com/sharing/share-offsite/?url=${enc}`; if(facebook) facebook.href=`https://www.facebook.com/sharer/sharer.php?u=${enc}`; } function openModal(id){ lastFocus=document.activeElement; hydrate(id); if(!isOpen()){ modal.removeAttribute('hidden'); } requestAnimationFrame(()=>{ input.focus(); }); trapFocus(); } function closeModal(){ if(!isOpen()) return; modal.setAttribute('hidden',''); if(lastFocus && typeof lastFocus.focus==='function') lastFocus.focus(); } function copyCurrent(){ try{ navigator.clipboard.writeText(input.value).then(()=>feedback(true),()=>fallback()); } catch(e){ fallback(); } } function fallback(){ input.select(); try{ document.execCommand('copy'); feedback(true);}catch(e){ feedback(false);} } function feedback(ok){ if(!copyBtn) return; const icon=copyBtn.querySelector('i'); if(!icon) return; const prev=copyBtn.getAttribute('data-prev')||icon.className; if(!copyBtn.getAttribute('data-prev')) copyBtn.setAttribute('data-prev',prev); icon.className= ok ? 'fa-duotone fa-clipboard-check':'fa-duotone fa-circle-exclamation'; setTimeout(()=>{ icon.className=prev; },1800); } function handleShareClick(e){ e.preventDefault(); const btn=e.currentTarget; const id=btn.getAttribute('data-share-target'); if(id) openModal(id); } function bindShareButtons(){ document.querySelectorAll('.h-share').forEach(btn=>{ if(!btn.dataset.hShareBound){ btn.addEventListener('click', handleShareClick); btn.dataset.hShareBound='1'; } }); } bindShareButtons(); if(document.readyState==='loading'){ document.addEventListener('DOMContentLoaded', bindShareButtons); } else { requestAnimationFrame(bindShareButtons); } document.addEventListener('click', function(e){ const shareBtn=e.target.closest && e.target.closest('.h-share'); if(shareBtn && !shareBtn.dataset.hShareBound){ handleShareClick.call(shareBtn, e); } }, true); document.addEventListener('click', e=>{ if(e.target===modal) closeModal(); if(e.target.closest && e.target.closest('.hsm-close')){ e.preventDefault(); closeModal(); } if(copyBtn && (e.target===copyBtn || (e.target.closest && e.target.closest('.hsm-copy')))) { e.preventDefault(); copyCurrent(); } }); document.addEventListener('keydown', e=>{ if(e.key==='Escape' && isOpen()) closeModal(); }); function trapFocus(){ if(trapBound) return; trapBound=true; modal.addEventListener('keydown', f=>{ if(f.key==='Tab' && isOpen()){ const focusable=[...modal.querySelectorAll('a[href],button,input,textarea,select,[tabindex]:not([tabindex="-1"])')].filter(el=>!el.hasAttribute('disabled')); if(!focusable.length) return; const first=focusable[0]; const last=focusable[focusable.length-1]; if(f.shiftKey && document.activeElement===first){ f.preventDefault(); last.focus(); } else if(!f.shiftKey && document.activeElement===last){ f.preventDefault(); first.focus(); } } }); } if(closeBtn) closeBtn.addEventListener('click', e=>{ e.preventDefault(); closeModal(); }); })(); </script>This guide teaches you how to analyze and optimize query performance in Geode using EXPLAIN, PROFILE, indexes, and the cost-based optimizer. You’ll learn to diagnose slow queries, understand execution plans, and apply optimization strategies. Performance Tuning Process: <ol> <li>Measure - Use PROFILE to identify slow queries</li> <li>Analyze - Use EXPLAIN to understand execution plans</li> <li>Optimize - Apply indexes, rewrite queries, tune configuration</li> <li>Verify - Re-measure to confirm improvements</li> </ol> Quick Links: <ul> <li><a href="https://geodedb.com/docs/query/performance-tuning/#using-explain" >Using EXPLAIN</a> </li> <li><a href="https://geodedb.com/docs/query/performance-tuning/#using-profile" >Using PROFILE</a> </li> <li><a href="https://geodedb.com/docs/query/performance-tuning/#index-optimization" >Index Optimization</a> </li> <li><a href="https://geodedb.com/docs/query/performance-tuning/#query-rewriting-patterns" >Query Rewriting Patterns</a> </li> <li><a href="https://geodedb.com/docs/query/performance-tuning/#common-performance-issues" >Common Performance Issues</a> </li> </ul> <hr> <h3 id="using-explain" class="position-relative d-flex align-items-center group"> Using EXPLAIN <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="using-explain" aria-haspopup="dialog" aria-label="Share link: Using EXPLAIN"> Share link </button> </h3>EXPLAIN shows the query execution plan without running the query. Use it to understand how Geode will execute your query before committing resources. <h4 id="basic-explain-syntax" class="position-relative d-flex align-items-center group"> Basic EXPLAIN Syntax <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="basic-explain-syntax" aria-haspopup="dialog" aria-label="Share link: Basic EXPLAIN Syntax"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN <statement> </code></pre></div> <h4 id="understanding-execution-plans" class="position-relative d-flex align-items-center group"> Understanding Execution Plans <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="understanding-execution-plans" aria-haspopup="dialog" aria-label="Share link: Understanding Execution Plans"> Share link </button> </h4>EXPLAIN output shows logical operators Geode will use: <table> <thead> <tr> <th>Operator</th> <th>Description</th> <th>Performance Impact</th> </tr> </thead> <tbody> <tr> <td><code>NodeScan</code></td> <td>Scans all nodes (or by label)</td> <td>O(n) - can be slow for large graphs</td> </tr> <tr> <td><code>IndexSeek</code></td> <td>Uses index for direct lookup</td> <td>O(log n) - fast</td> </tr> <tr> <td><code>IndexScan</code></td> <td>Scans index range</td> <td>O(k log n) - moderate</td> </tr> <tr> <td><code>ExpandRelationships</code></td> <td>Traverses relationships</td> <td>O(degree) per node</td> </tr> <tr> <td><code>Filter</code></td> <td>Applies WHERE predicates</td> <td>O(n) on input rows</td> </tr> <tr> <td><code>Project</code></td> <td>Selects/transforms columns</td> <td>O(n) - lightweight</td> </tr> <tr> <td><code>Sort</code></td> <td>Orders results (ORDER BY)</td> <td>O(n log n) - expensive</td> </tr> <tr> <td><code>Limit</code></td> <td>Restricts output rows</td> <td>O(1) - cheap</td> </tr> <tr> <td><code>Aggregate</code></td> <td>Groups and aggregates</td> <td>O(n) - moderate</td> </tr> </tbody> </table> <h4 id="explain-examples" class="position-relative d-flex align-items-center group"> EXPLAIN Examples <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="explain-examples" aria-haspopup="dialog" aria-label="Share link: EXPLAIN Examples"> Share link </button> </h4>Example 1: Simple Node Scan <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN MATCH (p:Person) RETURN p.name; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| plan | |-------------------| | EXPLAIN | | MATCH | | NodeScan | ← Scans all Person nodes | RETURN | | Project | ← Selects name property </code></pre></div>Analysis: Full node scan - acceptable for small graphs, slow for large. <hr> Example 2: Filtered Query <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN MATCH (p:Person) WHERE p.age > 25 RETURN p.name, p.age; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| plan | |-------------------| | EXPLAIN | | MATCH | | NodeScan | ← Scans all Person nodes | Filter | ← Filters age > 25 | RETURN | | Project | </code></pre></div>Analysis: Scans all nodes, then filters. Optimization needed if large dataset. <hr> Example 3: Relationship Traversal <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN MATCH (p:Person)-[:WORKS_FOR]->(c:Company) RETURN p.name, c.name; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| plan | |-------------------------| | EXPLAIN | | MATCH | | NodeScan | ← Start with Person nodes | ExpandRelationships | ← Traverse WORKS_FOR edges | RETURN | | Project | </code></pre></div>Analysis: Scans all Persons, expands relationships. Consider index on Company if filtering by company properties. <hr> Example 4: Complex Query with Aggregation <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN MATCH (p:Person)-[:WORKS_FOR]->(c:Company) WHERE p.age > 25 RETURN c.name, count(p) AS employee_count ORDER BY employee_count DESC LIMIT 10; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| plan | |-------------------------| | EXPLAIN | | MATCH | | NodeScan | | ExpandRelationships | | Filter | | RETURN | | Aggregate | ← Group by company, count employees | Sort | ← Sort by count (expensive!) | Limit | | Project | </code></pre></div>Analysis: Full scan → filter → aggregate → sort (expensive). Optimize filter with index. <hr> <h4 id="reading-explain-output" class="position-relative d-flex align-items-center group"> Reading EXPLAIN Output <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="reading-explain-output" aria-haspopup="dialog" aria-label="Share link: Reading EXPLAIN Output"> Share link </button> </h4>🚩 Red Flags (Slow Patterns): <ul> <li><code>NodeScan</code> without subsequent <code>Filter</code> on large graphs</li> <li><code>Sort</code> on large result sets</li> <li>Multiple <code>ExpandRelationships</code> without indexes</li> <li><code>Aggregate</code> on unindexed properties</li> </ul> ✅ Good Patterns (Fast): <ul> <li><code>IndexSeek</code> for direct lookups</li> <li><code>Filter</code> early in the plan (before expensive operations)</li> <li><code>Limit</code> without <code>Sort</code> for large datasets</li> <li>Index-backed <code>Aggregate</code> operations</li> </ul> <hr> <h3 id="using-profile" class="position-relative d-flex align-items-center group"> Using PROFILE <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="using-profile" aria-haspopup="dialog" aria-label="Share link: Using PROFILE"> Share link </button> </h3>PROFILE executes the query and returns real execution metrics. Use it to measure actual performance. <h4 id="basic-profile-syntax" class="position-relative d-flex align-items-center group"> Basic PROFILE Syntax <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="basic-profile-syntax" aria-haspopup="dialog" aria-label="Share link: Basic PROFILE Syntax"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE <statement> </code></pre></div> <h4 id="profile-metrics" class="position-relative d-flex align-items-center group"> PROFILE Metrics <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="profile-metrics" aria-haspopup="dialog" aria-label="Share link: PROFILE Metrics"> Share link </button> </h4><table> <thead> <tr> <th>Metric</th> <th>Description</th> </tr> </thead> <tbody> <tr> <td><code>rows_returned</code></td> <td>Number of rows returned</td> </tr> <tr> <td><code>columns</code></td> <td>Number of columns in result</td> </tr> <tr> <td><code>execution_time_ms</code></td> <td>Execution time in milliseconds</td> </tr> </tbody> </table> <h4 id="profile-examples" class="position-relative d-flex align-items-center group"> PROFILE Examples <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="profile-examples" aria-haspopup="dialog" aria-label="Share link: PROFILE Examples"> Share link </button> </h4>Example 1: Measure Simple Query <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person) RETURN p.name; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| metric | value | |---------------------|-------| | rows_returned | 42 | | columns | 1 | | execution_time_ms | 1 | </code></pre></div>Analysis: Returned 42 rows in 1ms - acceptable. <hr> Example 2: Measure Query with Filter <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person) WHERE p.age > 30 RETURN p.name, p.age; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| metric | value | |---------------------|-------| | rows_returned | 15 | | columns | 2 | | execution_time_ms | 2 | </code></pre></div>Analysis: 15 results in 2ms - good. If time is higher, add index on <code>age</code>. <hr> Example 3: Before/After Optimization Before (no index): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person) WHERE p.email = '[email protected]' RETURN p; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| metric | value | |---------------------|-------| | rows_returned | 1 | | execution_time_ms | 50 | ← Slow: full scan </code></pre></div>Create Index: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX person_email ON Person(email); </code></pre></div>After (with index): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person) WHERE p.email = '[email protected]' RETURN p; </code></pre></div>Output: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">| metric | value | |---------------------|-------| | rows_returned | 1 | | execution_time_ms | 2 | ← Fast: index seek! </code></pre></div>Result: Significant speedup with index. <hr> <h4 id="performance-measurement-workflow" class="position-relative d-flex align-items-center group"> Performance Measurement Workflow <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="performance-measurement-workflow" aria-haspopup="dialog" aria-label="Share link: Performance Measurement Workflow"> Share link </button> </h4>Step 1: Establish Baseline <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person)-[:KNOWS]->(friend:Person) WHERE p.age > 30 RETURN p.name, count(friend) AS friend_count ORDER BY friend_count DESC; </code></pre></div>Step 2: Analyze with EXPLAIN <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">EXPLAIN MATCH (p:Person)-[:KNOWS]->(friend:Person) WHERE p.age > 30 RETURN p.name, count(friend) AS friend_count ORDER BY friend_count DESC; </code></pre></div>Step 3: Apply Optimization (add index, rewrite query) <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX person_age ON Person(age); </code></pre></div>Step 4: Re-measure with PROFILE <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">PROFILE MATCH (p:Person)-[:KNOWS]->(friend:Person) WHERE p.age > 30 RETURN p.name, count(friend) AS friend_count ORDER BY friend_count DESC; </code></pre></div>Step 5: Compare Results <ul> <li>Before: execution_time_ms = 150</li> <li>After: execution_time_ms = 20</li> <li>Speedup: 7.5x ✅</li> </ul> <hr> <h3 id="index-optimization" class="position-relative d-flex align-items-center group"> Index Optimization <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="index-optimization" aria-haspopup="dialog" aria-label="Share link: Index Optimization"> Share link </button> </h3> <h4 id="index-types-in-geode" class="position-relative d-flex align-items-center group"> Index Types in Geode <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="index-types-in-geode" aria-haspopup="dialog" aria-label="Share link: Index Types in Geode"> Share link </button> </h4><table> <thead> <tr> <th>Index Type</th> <th>Use Case</th> <th>Creation Syntax</th> </tr> </thead> <tbody> <tr> <td>B-tree (default)</td> <td>Exact matches, range queries, sorting</td> <td><code>CREATE INDEX name ON Label(property)</code></td> </tr> <tr> <td>Hash</td> <td>Exact lookups only (faster than B-tree)</td> <td><code>CREATE INDEX name ON Label(property) USING hash</code></td> </tr> <tr> <td>Fulltext</td> <td>Text search, CONTAINS queries</td> <td><code>CREATE FULLTEXT INDEX name ON Label(property)</code></td> </tr> <tr> <td>Spatial</td> <td>Geographic queries (distance, bounding box)</td> <td><code>CREATE SPATIAL INDEX name ON Label(property)</code></td> </tr> <tr> <td>Vector</td> <td>Similarity search (embeddings, recommendations)</td> <td><code>CREATE VECTOR INDEX name ON Label(property)</code></td> </tr> </tbody> </table> <h4 id="creating-indexes" class="position-relative d-flex align-items-center group"> Creating Indexes <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="creating-indexes" aria-haspopup="dialog" aria-label="Share link: Creating Indexes"> Share link </button> </h4>B-tree Index (Default) <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- For exact matches and range queries CREATE INDEX person_age ON Person(age); CREATE INDEX product_price ON Product(price); -- Use for queries like: MATCH (p:Person) WHERE p.age > 30 RETURN p; MATCH (prod:Product) WHERE prod.price < 100 RETURN prod; </code></pre></div>Hash Index <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- For exact lookups only (faster than B-tree for equality) CREATE INDEX person_email ON Person(email) USING hash; -- Use for queries like: MATCH (p:Person {email: '[email protected]'}) RETURN p; MATCH (p:Person) WHERE p.email = '[email protected]' RETURN p; </code></pre></div>Fulltext Index (BM25-Optimized) <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- For text search with relevance ranking CREATE FULLTEXT INDEX article_content ON Article(content); CREATE FULLTEXT INDEX doc_abstract ON Document(abstract); -- Use for queries like: MATCH (a:Article) WHERE a.content CONTAINS 'graph database' RETURN a; MATCH (d:Document) WHERE d.abstract CONTAINS 'machine learning' RETURN d.title, bm25_score(d.abstract, 'machine learning') AS relevance ORDER BY relevance DESC; </code></pre></div>Spatial Index <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- For geographic queries CREATE SPATIAL INDEX location_coords ON Location(coordinates); -- Use for queries like: MATCH (loc:Location) WHERE distance(loc.coordinates, point(37.7749, -122.4194)) < 10000 -- 10km RETURN loc.name, distance(loc.coordinates, point(37.7749, -122.4194)) AS dist ORDER BY dist; </code></pre></div>Vector Index (HNSW) <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- For similarity search (embeddings) CREATE VECTOR INDEX doc_embedding ON Document(embedding); -- Use for queries like: MATCH (doc:Document) WHERE vector_cosine(doc.embedding, $query_embedding) > 0.7 RETURN doc.title, vector_cosine(doc.embedding, $query_embedding) AS similarity ORDER BY similarity DESC LIMIT 10; </code></pre></div> <h4 id="when-to-create-indexes" class="position-relative d-flex align-items-center group"> When to Create Indexes <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="when-to-create-indexes" aria-haspopup="dialog" aria-label="Share link: When to Create Indexes"> Share link </button> </h4>✅ Create Indexes For: <ol> <li> Frequently filtered properties <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Queries like WHERE p.status = 'active' are common CREATE INDEX user_status ON User(status); </code></pre></div></li> <li> Properties used in ORDER BY <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Queries like ORDER BY created_at DESC CREATE INDEX post_created ON Post(created_at); </code></pre></div></li> <li> Properties in JOIN conditions <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Foreign-key-like lookups CREATE INDEX order_customer_id ON Order(customer_id); </code></pre></div></li> <li> Unique identifiers <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Email, username, SKU, etc. CREATE INDEX user_email ON User(email) USING hash; CREATE INDEX product_sku ON Product(sku) USING hash; </code></pre></div></li> </ol> ❌ Don’t Create Indexes For: <ol> <li>Rarely queried properties - Index overhead not worth it</li> <li>High-cardinality properties with rare lookups - E.g., free-text descriptions</li> <li>Properties that change frequently - Index maintenance overhead</li> <li>Small datasets - Full scans are fast enough (<1000 nodes)</li> </ol> <h4 id="viewing-indexes" class="position-relative d-flex align-items-center group"> Viewing Indexes <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="viewing-indexes" aria-haspopup="dialog" aria-label="Share link: Viewing Indexes"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Show all indexes SHOW INDEXES; -- Output: | index_name | label | property | type | |------------------|---------|-------------|-----------| | person_email | Person | email | hash | | person_age | Person | age | btree | | article_content | Article | content | fulltext | | doc_embedding | Document| embedding | vector | </code></pre></div> <h4 id="dropping-indexes" class="position-relative d-flex align-items-center group"> Dropping Indexes <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="dropping-indexes" aria-haspopup="dialog" aria-label="Share link: Dropping Indexes"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Drop unused or redundant indexes DROP INDEX person_age; DROP INDEX article_content; </code></pre></div><hr> <h3 id="bm25-full-text-search-optimization" class="position-relative d-flex align-items-center group"> BM25 Full-Text Search Optimization <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="bm25-full-text-search-optimization" aria-haspopup="dialog" aria-label="Share link: BM25 Full-Text Search Optimization"> Share link </button> </h3>Geode’s cost-based optimizer automatically uses BM25 relevance ranking for fulltext searches, providing enterprise-grade search quality. <h4 id="bm25-scoring-formula" class="position-relative d-flex align-items-center group"> BM25 Scoring Formula <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="bm25-scoring-formula" aria-haspopup="dialog" aria-label="Share link: BM25 Scoring Formula"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-fallback" data-lang="fallback">score(q,d) = Σ IDF(qi) × f(qi,d) × (k1 + 1) / (f(qi,d) + k1 × (1 - b + b × |d| / avgdl)) </code></pre></div>Where: <ul> <li><code>IDF(qi)</code> = Inverse document frequency (rare terms score higher)</li> <li><code>f(qi,d)</code> = Term frequency in document</li> <li><code>|d|</code> = Document length in words</li> <li><code>avgdl</code> = Average document length in collection</li> <li><code>k1 = 1.2</code> = Term frequency saturation parameter</li> <li><code>b = 0.75</code> = Length normalization parameter</li> </ul> <h4 id="bm25-optimization-examples" class="position-relative d-flex align-items-center group"> BM25 Optimization Examples <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="bm25-optimization-examples" aria-haspopup="dialog" aria-label="Share link: BM25 Optimization Examples"> Share link </button> </h4>Example 1: Simple Text Search <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Create fulltext index CREATE FULLTEXT INDEX article_content ON Article(content); -- Query automatically uses BM25 MATCH (a:Article) WHERE a.content CONTAINS 'machine learning algorithms' RETURN a.title, a.author ORDER BY a.relevance_score DESC; </code></pre></div>Optimizer Behavior: <ul> <li>Base cost: 25.0 (vs 20.0 for standard fulltext)</li> <li>Query complexity: 1.0 + (3 terms - 1) × 0.3 = 1.6x multiplier</li> <li>Corpus scaling: Logarithmic with document count</li> <li>Result: Relevance-ranked results, 40-60% better quality</li> </ul> <hr> Example 2: Multi-Term Search with Ranking <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (doc:Document) WHERE doc.abstract CONTAINS 'artificial intelligence' AND doc.keywords CONTAINS 'neural networks' RETURN doc.title, bm25_score(doc.abstract, 'artificial intelligence neural networks') AS relevance ORDER BY relevance DESC LIMIT 10; </code></pre></div>Performance: <ul> <li>BM25 considers document length and term frequency</li> <li>Shorter documents with exact matches score higher</li> <li>Longer documents are penalized unless terms are frequent</li> <li>Typical speedup over naive text search: 2-5x with better results</li> </ul> <hr> Example 3: Corpus-Aware Optimization <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Technical documentation (high vocabulary density, long docs) MATCH (tech_doc:TechnicalDocument) WHERE tech_doc.content CONTAINS 'distributed systems architecture' RETURN tech_doc.title, tech_doc.complexity_score; -- Optimizer adjusts: -- - Vocabulary density > 100 → 1.2x cost multiplier -- - Average document length > 1000 words → length normalization -- - Result: Accurate cost estimation for complex corpus </code></pre></div><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Social media posts (low vocabulary, short docs) MATCH (post:SocialPost) WHERE post.text CONTAINS 'climate change' RETURN post.text, post.engagement_score ORDER BY post.timestamp DESC; -- Optimizer adjusts: -- - Vocabulary density < 20 → 0.9x cost multiplier (simpler) -- - Average document length < 200 words → reduced length penalty -- - Result: Faster execution for short-form content </code></pre></div> <h4 id="bm25-vs-standard-fulltext-comparison" class="position-relative d-flex align-items-center group"> BM25 vs Standard Fulltext Comparison <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="bm25-vs-standard-fulltext-comparison" aria-haspopup="dialog" aria-label="Share link: BM25 vs Standard Fulltext Comparison"> Share link </button> </h4><table> <thead> <tr> <th>Metric</th> <th>Standard Fulltext</th> <th>BM25 Enhanced</th> <th>Benefit</th> </tr> </thead> <tbody> <tr> <td>Base cost</td> <td>20.0</td> <td>25.0</td> <td>25% overhead for ranking</td> </tr> <tr> <td>Query complexity</td> <td>20% per term</td> <td>30% per term</td> <td>Better multi-term accuracy</td> </tr> <tr> <td>Corpus scaling</td> <td>Linear</td> <td>Logarithmic</td> <td>Better large-scale performance</td> </tr> <tr> <td>Search quality</td> <td>Term matching</td> <td>Relevance ranking</td> <td>40-60% better results</td> </tr> <tr> <td>Cost accuracy</td> <td>Heuristic</td> <td>Statistics-based</td> <td>25-35% more accurate</td> </tr> </tbody> </table> <hr> <h3 id="query-rewriting-patterns" class="position-relative d-flex align-items-center group"> Query Rewriting Patterns <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="query-rewriting-patterns" aria-haspopup="dialog" aria-label="Share link: Query Rewriting Patterns"> Share link </button> </h3> <h4 id="pattern-1-push-filters-early" class="position-relative d-flex align-items-center group"> Pattern 1: Push Filters Early <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="pattern-1-push-filters-early" aria-haspopup="dialog" aria-label="Share link: Pattern 1: Push Filters Early"> Share link </button> </h4>❌ Bad (Filter Late): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person)-[:KNOWS]->(friend:Person) WHERE p.age > 30 AND friend.age > 25 RETURN p.name, friend.name; </code></pre></div>✅ Good (Filter Early): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person {age: 30}) -- Filter immediately MATCH (p)-[:KNOWS]->(friend:Person) WHERE friend.age > 25 -- Filter on expansion RETURN p.name, friend.name; </code></pre></div>Why: Reduces nodes to expand from, fewer relationship traversals. <hr> <h4 id="pattern-2-avoid-cartesian-products" class="position-relative d-flex align-items-center group"> Pattern 2: Avoid Cartesian Products <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="pattern-2-avoid-cartesian-products" aria-haspopup="dialog" aria-label="Share link: Pattern 2: Avoid Cartesian Products"> Share link </button> </h4>❌ Bad (Cartesian Product): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) MATCH (c:Company) WHERE p.company_id = c.id RETURN p.name, c.name; -- If 1000 people × 100 companies = 100,000 intermediate rows! </code></pre></div>✅ Good (Connected Pattern): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person)-[:WORKS_AT]->(c:Company) RETURN p.name, c.name; -- Direct relationship traversal </code></pre></div>Why: Explicit relationships prevent combinatorial explosion. <hr> <h4 id="pattern-3-use-limits-early" class="position-relative d-flex align-items-center group"> Pattern 3: Use Limits Early <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="pattern-3-use-limits-early" aria-haspopup="dialog" aria-label="Share link: Pattern 3: Use Limits Early"> Share link </button> </h4>❌ Bad (Limit After Sort): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) RETURN p.name, p.age ORDER BY p.age DESC; -- Sorts ALL people -- Then application takes only first 10 </code></pre></div>✅ Good (Limit in Query): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) RETURN p.name, p.age ORDER BY p.age DESC LIMIT 10; -- Database returns only 10 </code></pre></div>Why: Database can use top-K optimization, avoid sorting/returning all rows. <hr> <h4 id="pattern-4-denormalize-for-read-performance" class="position-relative d-flex align-items-center group"> Pattern 4: Denormalize for Read Performance <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="pattern-4-denormalize-for-read-performance" aria-haspopup="dialog" aria-label="Share link: Pattern 4: Denormalize for Read Performance"> Share link </button> </h4>❌ Bad (Join on Every Query): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (order:Order)-[:CONTAINS]->(:LineItem)-[:OF_PRODUCT]->(product:Product) WHERE order.id = 12345 RETURN sum(product.price) AS total; -- Complex join every time </code></pre></div>✅ Good (Precompute Total): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Store total on Order when creating MATCH (order:Order {id: 12345}) RETURN order.total; -- Single property lookup -- Update total when items change (trade write speed for read speed) MATCH (order:Order {id: 12345})-[:CONTAINS]->(li:LineItem) SET order.total = sum(li.subtotal); </code></pre></div>Why: Read-heavy workloads benefit from denormalization. <hr> <h4 id="pattern-5-use-optional-match-sparingly" class="position-relative d-flex align-items-center group"> Pattern 5: Use OPTIONAL MATCH Sparingly <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="pattern-5-use-optional-match-sparingly" aria-haspopup="dialog" aria-label="Share link: Pattern 5: Use OPTIONAL MATCH Sparingly"> Share link </button> </h4>❌ Bad (Multiple OPTIONAL): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) OPTIONAL MATCH (p)-[:WORKS_AT]->(c:Company) OPTIONAL MATCH (p)-[:LIVES_IN]->(city:City) OPTIONAL MATCH (p)-[:STUDIED_AT]->(school:School) RETURN p, c, city, school; -- Many optional branches slow execution </code></pre></div>✅ Good (Filter Required First): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- If most people work somewhere, make it required MATCH (p:Person)-[:WORKS_AT]->(c:Company) OPTIONAL MATCH (p)-[:LIVES_IN]->(city:City) RETURN p, c, city; </code></pre></div>Why: Required MATCH is faster than OPTIONAL. <hr> <h3 id="common-performance-issues" class="position-relative d-flex align-items-center group"> Common Performance Issues <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="common-performance-issues" aria-haspopup="dialog" aria-label="Share link: Common Performance Issues"> Share link </button> </h3> <h4 id="issue-1-full-node-scans" class="position-relative d-flex align-items-center group"> Issue 1: Full Node Scans <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="issue-1-full-node-scans" aria-haspopup="dialog" aria-label="Share link: Issue 1: Full Node Scans"> Share link </button> </h4>Symptom: Queries slow on large graphs, EXPLAIN shows <code>NodeScan</code>. Example: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) WHERE p.email = '[email protected]' RETURN p; -- EXPLAIN shows: NodeScan → Filter </code></pre></div>Solution: Create index on filtered property. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX person_email ON Person(email); -- Now EXPLAIN shows: IndexSeek </code></pre></div>Impact: Significant improvement on large datasets. <hr> <h4 id="issue-2-expensive-sorts" class="position-relative d-flex align-items-center group"> Issue 2: Expensive Sorts <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="issue-2-expensive-sorts" aria-haspopup="dialog" aria-label="Share link: Issue 2: Expensive Sorts"> Share link </button> </h4>Symptom: Queries with ORDER BY are slow, high memory usage. Example: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) RETURN p.name, p.created_at ORDER BY p.created_at DESC; -- Sorts ALL people in memory </code></pre></div>Solution 1: Create index on sort property. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX person_created ON Person(created_at); -- Index provides sorted iteration </code></pre></div>Solution 2: Add LIMIT if you don’t need all results. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) RETURN p.name, p.created_at ORDER BY p.created_at DESC LIMIT 100; -- Only sort top 100 </code></pre></div>Impact: 5-50x speedup depending on dataset size. <hr> <h4 id="issue-3-supernode-expansion" class="position-relative d-flex align-items-center group"> Issue 3: Supernode Expansion <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="issue-3-supernode-expansion" aria-haspopup="dialog" aria-label="Share link: Issue 3: Supernode Expansion"> Share link </button> </h4>Symptom: Queries traversing high-degree nodes are slow. Example: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- "Popular" node with 1,000,000 followers MATCH (celebrity:Person {name: "Celebrity"})-[:FOLLOWED_BY]->(follower:Person) RETURN follower.name; -- Expands 1M relationships! </code></pre></div>Solution 1: Add LIMIT and pagination. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (celebrity:Person {name: "Celebrity"})-[:FOLLOWED_BY]->(follower:Person) RETURN follower.name ORDER BY follower.id SKIP 1000 LIMIT 100; </code></pre></div>Solution 2: Denormalize with cached counts. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (celebrity:Person {name: "Celebrity"}) RETURN celebrity.follower_count; -- Precomputed property </code></pre></div>Solution 3: Partition relationships. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Instead of one celebrity node with 1M edges: -- Create follower buckets (celebrity)-[:HAS_FOLLOWER_BUCKET]->(bucket:FollowerBucket {range: "A-F"}) (bucket)-[:CONTAINS_FOLLOWER]->(follower:Person) </code></pre></div>Impact: Avoid O(n) expansion on supernodes. <hr> <h4 id="issue-4-inefficient-aggregations" class="position-relative d-flex align-items-center group"> Issue 4: Inefficient Aggregations <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="issue-4-inefficient-aggregations" aria-haspopup="dialog" aria-label="Share link: Issue 4: Inefficient Aggregations"> Share link </button> </h4>Symptom: COUNT, SUM, AVG queries are slow. Example: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (p:Person) RETURN count(p); -- Scans all Person nodes </code></pre></div>Solution 1: Maintain counts in metadata node. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (stats:Statistics {type: "Person"}) RETURN stats.total_count; -- Update on create/delete MATCH (stats:Statistics {type: "Person"}) SET stats.total_count = stats.total_count + 1; </code></pre></div>Solution 2: Use index-backed aggregation (if supported). <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX person_label ON Person(label); -- Future: Optimizer uses index stats for count </code></pre></div>Impact: Constant time vs O(n) for large datasets. <hr> <h4 id="issue-5-unindexed-joins" class="position-relative d-flex align-items-center group"> Issue 5: Unindexed JOINs <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="issue-5-unindexed-joins" aria-haspopup="dialog" aria-label="Share link: Issue 5: Unindexed JOINs"> Share link </button> </h4>Symptom: Queries joining on properties without indexes are slow. Example: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (order:Order), (customer:Customer) WHERE order.customer_id = customer.id RETURN order.id, customer.name; -- Cartesian product: orders × customers! </code></pre></div>Solution 1: Use explicit relationships. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">MATCH (order:Order)-[:PLACED_BY]->(customer:Customer) RETURN order.id, customer.name; </code></pre></div>Solution 2: If relationships don’t exist, create indexes. <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">CREATE INDEX order_customer_id ON Order(customer_id); CREATE INDEX customer_id ON Customer(id); </code></pre></div>Impact: Avoids expensive Cartesian products. <hr> <h3 id="performance-checklist" class="position-relative d-flex align-items-center group"> Performance Checklist <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="performance-checklist" aria-haspopup="dialog" aria-label="Share link: Performance Checklist"> Share link </button> </h3> <h4 id="before-query-execution" class="position-relative d-flex align-items-center group"> Before Query Execution <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="before-query-execution" aria-haspopup="dialog" aria-label="Share link: Before Query Execution"> Share link </button> </h4><ul> <li><input disabled="" type="checkbox"> Use EXPLAIN to review execution plan</li> <li><input disabled="" type="checkbox"> Check for NodeScan - add indexes if needed</li> <li><input disabled="" type="checkbox"> Verify index usage - IndexSeek is ideal</li> <li><input disabled="" type="checkbox"> Look for expensive sorts - consider index on ORDER BY property</li> <li><input disabled="" type="checkbox"> Identify Cartesian products - use relationships instead</li> </ul> <h4 id="after-query-execution" class="position-relative d-flex align-items-center group"> After Query Execution <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="after-query-execution" aria-haspopup="dialog" aria-label="Share link: After Query Execution"> Share link </button> </h4><ul> <li><input disabled="" type="checkbox"> Use PROFILE to measure actual performance</li> <li><input disabled="" type="checkbox"> Compare execution_time_ms to expectations</li> <li><input disabled="" type="checkbox"> Check rows_returned - is it what you expected?</li> <li><input disabled="" type="checkbox"> Verify index hit - did optimizer use your index?</li> <li><input disabled="" type="checkbox"> Re-measure after optimization - confirm improvement</li> </ul> <h4 id="index-strategy" class="position-relative d-flex align-items-center group"> Index Strategy <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="index-strategy" aria-haspopup="dialog" aria-label="Share link: Index Strategy"> Share link </button> </h4><ul> <li><input disabled="" type="checkbox"> Index frequently filtered properties</li> <li><input disabled="" type="checkbox"> Index ORDER BY properties</li> <li><input disabled="" type="checkbox"> Use hash indexes for exact lookups</li> <li><input disabled="" type="checkbox"> Use fulltext indexes for text search</li> <li><input disabled="" type="checkbox"> Use vector indexes for similarity search</li> <li><input disabled="" type="checkbox"> Don’t over-index - each index adds write overhead</li> </ul> <hr> <h3 id="advanced-optimization-techniques" class="position-relative d-flex align-items-center group"> Advanced Optimization Techniques <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="advanced-optimization-techniques" aria-haspopup="dialog" aria-label="Share link: Advanced Optimization Techniques"> Share link </button> </h3> <h4 id="materialized-views" class="position-relative d-flex align-items-center group"> Materialized Views <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="materialized-views" aria-haspopup="dialog" aria-label="Share link: Materialized Views"> Share link </button> </h4>For expensive aggregations run frequently, precompute results: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-gql" data-lang="gql">-- Instead of computing on every query: MATCH (p:Person)-[:WORKS_FOR]->(c:Company {name: "Acme"}) RETURN count(p) AS employee_count; -- Maintain a cached value: MATCH (c:Company {name: "Acme"}) RETURN c.employee_count; -- Updated on hire/termination </code></pre></div> <h4 id="query-result-caching" class="position-relative d-flex align-items-center group"> Query Result Caching <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="query-result-caching" aria-haspopup="dialog" aria-label="Share link: Query Result Caching"> Share link </button> </h4>Cache frequent query results at application level: <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-python" data-lang="python">from functools import lru_cache @lru_cache(maxsize=128) def get_company_employees(company_name): return db.query( "MATCH (p:Person)-[:WORKS_FOR]->(c:Company {name: $name}) RETURN p", {"name": company_name} ) </code></pre></div> <h4 id="batch-operations" class="position-relative d-flex align-items-center group"> Batch Operations <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="batch-operations" aria-haspopup="dialog" aria-label="Share link: Batch Operations"> Share link </button> </h4>Instead of N queries, use one batched query: ❌ Bad (N+1 Query Problem): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-python" data-lang="python">for order_id in order_ids: # 1000 orders result = db.query("MATCH (o:Order {id: $id}) RETURN o", {"id": order_id}) # 1000 individual queries! </code></pre></div>✅ Good (Batch Query): <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-python" data-lang="python">result = db.query( "MATCH (o:Order) WHERE o.id IN $ids RETURN o", {"ids": order_ids} ) # Single query </code></pre></div><hr> <h3 id="monitoring-query-performance" class="position-relative d-flex align-items-center group"> Monitoring Query Performance <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="monitoring-query-performance" aria-haspopup="dialog" aria-label="Share link: Monitoring Query Performance"> Share link </button> </h3> <h4 id="enable-query-logging" class="position-relative d-flex align-items-center group"> Enable Query Logging <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="enable-query-logging" aria-haspopup="dialog" aria-label="Share link: Enable Query Logging"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-yaml" data-lang="yaml"># geode.yaml logging: level: 'info' slow_query_threshold_ms: 1000 # Log queries > 1s </code></pre></div> <h4 id="prometheus-metrics" class="position-relative d-flex align-items-center group"> Prometheus Metrics <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="prometheus-metrics" aria-haspopup="dialog" aria-label="Share link: Prometheus Metrics"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-bash" data-lang="bash"># Query duration histogram geode_query_duration_seconds_bucket{le="0.1"} geode_query_duration_seconds_bucket{le="1.0"} geode_query_duration_seconds_bucket{le="10.0"} # Query rate rate(geode_queries_total[5m]) # 95th percentile latency histogram_quantile(0.95, rate(geode_query_duration_seconds_bucket[5m])) </code></pre></div> <h4 id="grafana-dashboard-queries" class="position-relative d-flex align-items-center group"> Grafana Dashboard Queries <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="grafana-dashboard-queries" aria-haspopup="dialog" aria-label="Share link: Grafana Dashboard Queries"> Share link </button> </h4><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-promql" data-lang="promql"># Top 5 slowest query types topk(5, avg(rate(geode_query_duration_seconds_sum[5m])) by (query_type)) # Query throughput sum(rate(geode_queries_total[5m])) # Error rate rate(geode_query_errors_total[5m]) / rate(geode_queries_total[5m]) </code></pre></div><hr> <h3 id="next-steps" class="position-relative d-flex align-items-center group"> Next Steps <button type="button" class="h-share btn btn-link p-0 text-decoration-none link-secondary opacity-50 hover-opacity-100 transition-all ms-1" data-share-target="next-steps" aria-haspopup="dialog" aria-label="Share link: Next Steps"> Share link </button> </h3><ul> <li><a href="/docs/query/indexing-and-optimization/" >Indexing Guide</a> - Complete index reference</li> <li><a href="/docs/model/data-model-and-types/" >Data Model</a> - Type system and schema design</li> <li><a href="/docs/guides/troubleshooting/" >Troubleshooting</a> - Diagnose performance issues</li> <li><a href="/docs/configuration/server-configuration/" >Configuration</a> - Tune server performance</li> <li><a href="/docs/ops/observability/" >Monitoring</a> - Set up performance monitoring</li> </ul>